公开记录
软件健康报告模式 0.27.0 · 指标 1.13.0 · 2026-07-23 09:25 UTC

Oaklight / llm-rosetta

Production-ready LLM API translation layer for Python — bidirectional conversion between OpenAI, Anthropic & Google formats via hub-and-spoke IR. Optional API gateway. Streaming & non-streaming. Zero core deps. Contributions welcome!

PythonMIT★ 58 星标⑂ 5 复刻始于 2025年12月在 GitHub 上查看 ↗

Oaklight/llm-rosetta 的健康指数为 100 分中的 68 分,处于「中等」区间。 其得分最高的类别是Vitality(86/100),最低的是Community & Adoption(53/100)。 最近一次更新在 1 天前。 近期的大部分工作由 1 位贡献者完成。

68
总分 / 100
中等

软件健康指数

指标归入加权类别,统一采用 1–100 量表。总体分先取类别加权平均;当公开证据触发高风险司法辖区政策时,评级会按政策调整,并设置 49(有风险)的上限。AI 就绪度不计入总体分。

68
优秀85-100堪称典范;基本满足所有检验标准
良好70-84健康;仅有轻微不足
中等50-69可接受,但存在明显不足;建议进行审查
存在风险30-49存在重大薄弱环节;采用时应保持审慎
危急1-29问题严重(项目被弃置、仅有单一维护者、缺乏基本工程规范)
活力社区与采用可持续性与治理工程质量安全AI 就绪度

评分画像

每条轴代表一个类别。形状比平均值更重要——健康的对象会填满整个图形,而“一峰一谷”式画像意味着某一维度的优势正掩盖另一维度的风险。

所有权

Peng Ding个人账户
78 关注者104 个公开仓库始于 2015年6月University of Chicago

该仓库由个人账户拥有。相较于组织支持的项目,单一所有者项目的延续性风险更高。

软件包生态系统

注册表软件包版本月下载量版本数最近发布标签
PyPIllm-rosetta0.7.24,817363 天前openaianthropicgeminigoogle-genaiapi-adapterllm-gatewayllmirstreamingtool-callingmulti-providerresponses-api

按类别列示的指标

活力

项目是否仍有生命——是否仍在编写代码,是否仍在发布版本?

86优秀 · 占总体的 22%
评分方式
36/36推送新近度 — 最近一次推送于 1 天前
18.7/36提交节奏 — 52 周中有 27 周有提交
18/18提交量 — 最近一年 715 次提交
10/10OpenSSF Scorecard:Maintained — 30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10
所用输入
commits_last_year715
human_commit_share0.93
days_since_last_push1
active_weeks_last_year27

发布纪律

90优秀
评分方式
27/27有发布版本 — 已发布 36 个发布版本
36/36发布时效 — 最近一次发布版本于 3 天前
27/27发布节奏 — 约每 3.5 天发布一次
0/10OpenSSF Scorecard:Signed-Releases — Project has not signed or included provenance with any releases.
所用输入
releases_count36
latest_release_tagv0.7.2
releases_from_tags
days_since_latest_release3
mean_days_between_releases3.5

社区与采用

项目是否拥有用户、下载量与关注度,并具备欢迎贡献者参与的配置?

53中等 · 占总体的 18%

流行度与采用

34存在风险
评分方式
28.5/60星标 — 58 个星标
5/25复刻 — 5 个复刻
0/15关注者 — 1 位关注者
所用输入
forks5
stars58
watchers1
growth_stateunverified
growth_factor_pct100
growth_unverified_reasonno_history

社区健康

70良好
评分方式
22.5/22.5README
22.5/22.5许可证 — 可识别的许可证(MIT)
18/18CONTRIBUTING 指南
0/13.5行为准则
0/7.2议题模板
0/6.3PR 模板
所用输入
has_readme
has_license
has_contributing
has_issue_template
has_code_of_conduct
has_pull_request_template
评分方式
49.1/80月度下载量 — pypi 合计每月 4,817 次下载
0/20注册表被依赖数 — 该生态系统不报告此项
所用输入
packagesllm-rosetta
dependents
ecosystemspypi
total_downloads
monthly_downloads4,817
已排除计分(无数据或不适用):注册表被依赖数。 其余权重已重新归一化。

可持续性与治理

项目能否在其成员之外延续——巴士系数、响应能力、由谁支持,以及软件包的维护状况?

58中等 · 占总体的 24%
评分方式
9/54巴士系数 — 1 位贡献者贡献了半数提交
0.1/22.5提交分布 — 头号贡献者编写了 100% 的提交
2.7/13.5贡献者广度 — 2 位贡献者
3/10OpenSSF Scorecard:Contributors — project has 1 contributing companies or organizations -- score normalized to 3
所用输入
bus_factor1
contributors_sampled2
top_contributor_share0.997
评分方式
38.8/46.8议题解决 — 83% 的议题已关闭
33.9/38.3PR 接受 — 已裁定的 PR 中 186/210 已合并
1.5/15OpenSSF Scorecard:Code-Review — Found 2/20 approved changesets -- score normalized to 1
所用输入
merged_prs186
open_issues29
closed_issues143
issue_closed_ratio0.831
closed_unmerged_prs24
评分方式
10/30所有权背书 — 个人(用户)账户
0/20已验证域名 — 不适用于个人账户
13.6/25所有者影响力 — Oaklight 有 78 位关注者
25/25既往记录 — 104 个公开仓库,账户约 11 年
所用输入
followers78
owner_typeUser
is_verified
owner_loginOaklight
public_repos104
account_age_days4,043
已排除计分(无数据或不适用):已验证域名。 其余权重已重新归一化。
评分方式
25/25已发布且可解析 — pypi 上有 1 个软件包
35/35发布时效 — 最近一次发布于 3 天前
20/20版本历史 — 36 个已发布版本
20/20未被弃用 — 活跃,未被弃用或撤回
所用输入
packagesllm-rosetta
ecosystemspypi
any_deprecated
min_days_since_publish3

工程质量

基础的工程与文档实践是否到位?

82良好 · 占总体的 20%

工程实践

94优秀
评分方式
24/24CI 工作流 — 6 个工作流
24/24存在测试
16/16Linter 配置
9.6/9.6Pre-commit 钩子
0/6.4.editorconfig
20/20OpenSSF Scorecard:CI-Tests — 7 out of 7 merged PRs checked by a CI test -- score normalized to 10
所用输入
has_ci
has_tests
has_editorconfig
has_linter_config
has_precommit_config

文档

65中等
评分方式
30/30README
0/25文档目录
15/15文档 / 主页站点 — https://llm-rosetta.readthedocs.io
10/10仓库描述
10/10主题标签 — 14 个主题标签
0/10Wiki
所用输入
topicsanthropic, gemini, openai, python, streaming, anthropic-api, api-adapter, llm-gateway, openai-api, llmir, google-genai, multi-provider, responses-api, tool-calling
has_wiki
homepagehttps://llm-rosetta.readthedocs.io
has_readme
has_docs_dir
has_description

安全

可见的安全与供应链实践是否稳固,且不存在未解决的高风险司法辖区暴露?

59中等 · 占总体的 16%

安全态势

49存在风险
评分方式
7.5/7.5Binary-Artifacts — no binaries found in the repo
0/7.5Branch-Protection — branch protection not enabled on development/release branches
2.5/2.5CI-Tests — 7 out of 7 merged PRs checked by a CI test -- score normalized to 10
0/2.5CII-Best-Practices — no effort to earn an OpenSSF best practices badge detected
0.8/7.5Code-Review — Found 2/20 approved changesets -- score normalized to 1
0.8/2.5Contributors — project has 1 contributing companies or organizations -- score normalized to 3
10/10Dangerous-Workflow — no dangerous workflow patterns detected
7.5/7.5Dependency-Update-Tool — update tool detected
0/5Fuzzing — project is not fuzzed
2.5/2.5许可证 — license file detected
7.5/7.5Maintained — 30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10
5/5Packaging — packaging workflow detected
0/5Pinned-Dependencies — dependency not pinned by hash detected -- score normalized to 0
0/5SAST — SAST tool is not run on all commits -- score normalized to 0
0/5Security-Policy — security policy file not detected
0/7.5Signed-Releases — Project has not signed or included provenance with any releases.
0/7.5Token-Permissions — detected GitHub workflow tokens with excessive permissions
7.5/7.5Vulnerabilities — 0 existing vulnerabilities detected
所用输入
sourceopenssf_scorecard
checks_evaluated18
scorecard_versionv5.5.0
checks_inconclusive0
scorecard_aggregate4.9
评分方式
35/35直接依赖不含已知公告 — 没有直接依赖携带已知公告
0/25间接依赖不含已知公告 — 在此范围内,传递依赖集合无法与开发和测试依赖区分
0/40没有长期未处理的公告 — 没有公告带有发布日期
所用输入
sourceosv
advisories6
affected_packages2
assessed_packages82
unassessed_packages13
affected_by_severityhigh 2
direct_affected_packages0
已排除计分(无数据或不适用):间接依赖不含已知公告, 没有长期未处理的公告。 其余权重已重新归一化。 已将 82 个已解析依赖与 OSV 比对。 有 13 项无法评估——没有已解析的版本、生态系统不受支持,或超出所报告的软件包清单。 该仓库未发布任何索引可解析的软件包,因此改为评估仓库依赖图。该图将开发与测试版本固定同交付的依赖混在一起,因此仅对声明的运行时依赖计分;传递性发现仅作为背景信息列出,不计入评分。 未对可达性进行分析。

AI 就绪度

该仓库在多大程度上具备与 AI 编码代理协同开发与维护的条件?这是一枚独立的实验性徽章——权重为 0.0,因此单独呈现,不影响总体健康评分。

74良好 · 占总体的 0%
评分方式
45/45代理指令 — AGENTS.md, CLAUDE.md
0/15机器可读文档(llms.txt)
40/40可读的提交历史 — 93 次人类提交中有 91 次说明了意图(结构化标题或解释性正文)
所用输入
has_llms_txt
legible_history_share0.978
agent_instruction_filesAGENTS.md, CLAUDE.md
agent_instruction_max_bytes11,571
评分方式
18/18一条命令的引导启动 — Makefile
22/22自动化测试
11/11Lint / 格式化配置
11/11静态类型检查 — src/llm_rosetta/py.typed
10/10可复现环境 — Dockerfile
0/10已体现的代理实践 — 最近 100 次提交中没有代理编写的提交
5/8自动化维护 — 已配置依赖自动化,但在抽样提交中未观察到
0/10OpenSSF Scorecard:Pinned-Dependencies — dependency not pinned by hash detected -- score normalized to 0
所用输入
has_nix
has_tests
lockfiles
has_dockerfile
typed_language
bootstrap_filesMakefile
has_devcontainer
has_linter_config
typecheck_configssrc/llm_rosetta/py.typed
agent_commit_share0
toolchain_manifests
dependency_bot_commit_share0
评分方式
27/45可类型检查的代码 — Python,已配置类型检查(src/llm_rosetta/py.typed)
54.3/55可控的文件大小 — 采样的 297 个源文件中有 4 个超过 60KB
所用输入
primary_languagePython
largest_source_bytes91,915
source_files_sampled297
oversized_source_files4

机器可读接口

40存在风险
评分方式
0/40API 模式(OpenAPI/GraphQL/proto)
0/20MCP 服务器
40/40可运行示例 — examples
所用输入
example_dirsexamples
has_mcp_signal
api_schema_files

关键数据

58GitHub 星标
2贡献者
715最近 12 个月提交数
1距最近推送天数
36发布版本数
1巴士系数(bus factor)
29开放议题
PyPI软件包生态系统数

数据采集警告

  • Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token
  • deps.dev does not index pypi:llm-rosetta@0.7.2; advisories assessed against the repository dependency graph instead

更多细节

Star 与 Fork 历史 0 ★ / 5 ⇿
0Star
5Fork
30发布

每颗 star 和每个 fork 的添加时间,来自 GitHub 并按天汇总。累计增长位于其构成来源——每日新增——的正上方,二者可相互对照:稳定的自然增长与短暂的突增形态截然不同。当这一差别可被衡量时,它会作为增长真实性予以报告。

12345512026-032026-052026-07
主版本 0次版本 4修订 23
OpenSSF Scorecard 4.9 / 10
4.9综合

来自开源项目 OpenSSF Scorecard 的独立、工具无关的安全评估。每项检查奖励的是安全实践本身,而非特定供应商的工具。Scorecard 无法判定的检查项标记为 不适用,并从安全评分中剔除(绝不按零分计)。Scorecard v5.5.0 · 2026-07-23 09:25 UTC

10Binary-Artifactsno binaries found in the repo
0Branch-Protectionbranch protection not enabled on development/release branches
10CI-Tests7 out of 7 merged PRs checked by a CI test -- score normalized to 10
0CII-Best-Practicesno effort to earn an OpenSSF best practices badge detected
1Code-ReviewFound 2/20 approved changesets -- score normalized to 1
3Contributorsproject has 1 contributing companies or organizations -- score normalized to 3
10Dangerous-Workflowno dangerous workflow patterns detected
10Dependency-Update-Toolupdate tool detected
0Fuzzingproject is not fuzzed
10Licenselicense file detected
10Maintained30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10
10Packagingpackaging workflow detected
0Pinned-Dependenciesdependency not pinned by hash detected -- score normalized to 0
0SASTSAST tool is not run on all commits -- score normalized to 0
0Security-Policysecurity policy file not detected
0Signed-ReleasesProject has not signed or included provenance with any releases.
0Token-Permissionsdetected GitHub workflow tokens with excessive permissions
10Vulnerabilities0 existing vulnerabilities detected
直接依赖 1
注册表软件包版本约束清单文件
PyPItyping_extensions>=4.0.0pyproject.toml
全部依赖 95

来自 GitHub 依赖图的完整解析依赖集合:2 个直接依赖与 93 个间接(传递)软件包。仓库提交锁文件时,传递闭包才是完整的。

注册表软件包版本关系
PyPItyping-extensions直接
PyPItyping-extensions4.15.0直接
PyPIannotated-doc0.0.4间接
PyPIannotated-types0.7.0间接
PyPIanthropic间接
PyPIanthropic0.105.2间接
PyPIanyio4.13.0间接
PyPIbackports-tarfile1.2.0间接
PyPIbuild间接
PyPIbuild1.5.0间接
PyPIcertifi2026.5.20间接
PyPIcffi2.0.0间接
PyPIcfgv3.5.0间接
PyPIcharset-normalizer3.4.7间接
PyPIcolorama0.4.6间接
PyPIcomplexipy间接
PyPIcomplexipy5.5.0间接
PyPIcoverage7.14.1间接
PyPIcryptography48.0.0间接
PyPIdistlib0.4.1间接
PyPIdistro1.9.0间接
PyPIdocstring-parser0.18.0间接
PyPIdocutils0.23间接
PyPIexceptiongroup1.3.1间接
PyPIfilelock3.29.1间接
PyPIgoogle-auth2.53.0间接
PyPIgoogle-genai间接
PyPIgoogle-genai2.8.0间接
PyPIh110.16.0间接
PyPIhttpcore1.0.9间接
PyPIhttpx间接
PyPIhttpx0.28.1间接
PyPIid1.6.1间接
PyPIidentify2.6.19间接
PyPIidna3.18间接
PyPIimportlib-metadata9.0.0间接
PyPIiniconfig2.3.0间接
PyPIjaraco-classes3.4.0间接
PyPIjaraco-context6.1.2间接
PyPIjaraco-functools4.5.0间接
PyPIjeepney0.9.0间接
PyPIjiter0.15.0间接
PyPIkeyring25.7.0间接
PyPIllm-rosetta间接
PyPImarkdown-it-py4.2.0间接
PyPImdurl0.1.2间接
PyPImore-itertools11.1.0间接
PyPInh30.3.5间接
PyPInodeenv1.10.0间接
PyPIopenai间接
PyPIopenai2.41.0间接
PyPIpackaging26.2间接
PyPIplatformdirs4.10.0间接
PyPIpluggy1.6.0间接
PyPIpre-commit间接
PyPIpre-commit4.6.0间接
PyPIpyasn10.6.3间接
PyPIpyasn1-modules0.4.2间接
PyPIpycparser3.0间接
PyPIpydantic2.13.4间接
PyPIpydantic-core2.46.4间接
PyPIpygments2.20.0间接
PyPIpyinstrument间接
PyPIpyproject-hooks1.2.0间接
PyPIpytest9.0.3间接
PyPIpytest-cov7.1.0间接
PyPIpython-discovery1.4.0间接
PyPIpython-dotenv间接
PyPIpython-dotenv1.2.2间接
PyPIpywin32-ctypes0.2.3间接
PyPIpyyaml6.0.3间接
PyPIreadme-renderer44.0间接
PyPIrequests2.34.2间接
PyPIrequests-toolbelt1.0.0间接
PyPIrfc39862.0.0间接
PyPIrich15.0.0间接
PyPIruff0.15.16间接
PyPIruff0.15.20间接
PyPIsecretstorage3.5.0间接
PyPIsetuptools间接
PyPIshellingham1.5.4间接
PyPIsniffio1.3.1间接
PyPItenacity9.1.4间接
PyPItomli2.4.1间接
PyPItqdm4.67.3间接
PyPItwine间接
PyPItwine6.2.0间接
PyPIty0.0.34间接
PyPIty0.0.54间接
PyPItyper0.26.7间接
PyPItyping-inspection0.4.2间接
PyPIurllib32.7.0间接
PyPIvirtualenv21.4.2间接
PyPIwebsockets16.0间接
PyPIzipp4.1.0间接
依赖安全公告 2

该仓库未发布可被索引解析的包,因此评估的是其自身的依赖图——共 82 个包,其中也包含从不交付的开发与测试版本固定:2 个存在已知公告,0 个为直接依赖。 有 13 个无法评估——没有已解析的版本、生态系统不受支持,或不在所列包清单之内。

软件包版本关系严重程度公告数修复版本
cryptography48.0.0间接148.0.1
pyasn10.6.3间接50.6.4

公告表示依赖图中记录的版本落入某条公告的受影响范围。可达性未经分析,且依赖图包含开发与测试的版本固定——某项发现可能只涉及工具链而非交付的软件。

原始 JSON 报告 机器可读
{
  "data": {
    "repo": {
      "topics": [
        "anthropic",
        "gemini",
        "openai",
        "python",
        "streaming",
        "anthropic-api",
        "api-adapter",
        "llm-gateway",
        "openai-api",
        "llmir",
        "google-genai",
        "multi-provider",
        "responses-api",
        "tool-calling"
      ],
      "is_fork": false,
      "size_kb": 4620,
      "has_wiki": false,
      "homepage": "https://llm-rosetta.readthedocs.io",
      "languages": {
        "HTML": 263360,
        "Shell": 8988,
        "Python": 3010781,
        "Makefile": 9669,
        "Dockerfile": 2729
      },
      "pushed_at": "2026-07-22T05:38:30Z",
      "created_at": "2025-12-09T04:00:30Z",
      "owner_type": "User",
      "updated_at": "2026-07-21T14:35:56Z",
      "description": "Production-ready LLM API translation layer for Python — bidirectional conversion between OpenAI, Anthropic & Google formats via hub-and-spoke IR. Optional API gateway. Streaming & non-streaming. Zero core deps. Contributions welcome!",
      "is_archived": false,
      "is_disabled": false,
      "license_spdx": "MIT",
      "default_branch": "master",
      "license_spdx_raw": "MIT",
      "primary_language": "Python",
      "significant_languages": [
        "Python"
      ]
    },
    "owner": {
      "blog": "https://oaklight.github.io/",
      "name": "Peng Ding",
      "type": "User",
      "login": "Oaklight",
      "company": "University of Chicago",
      "location": null,
      "followers": 78,
      "avatar_url": "https://avatars.githubusercontent.com/u/13076552?v=4",
      "created_at": "2015-06-27T11:56:41Z",
      "is_verified": null,
      "public_repos": 104,
      "account_age_days": 4043
    },
    "license": {
      "state": "standard",
      "spdx_id": "MIT",
      "raw_spdx": "MIT",
      "file_present": true,
      "scorecard_found": true,
      "profile_has_license": true
    },
    "activity": {
      "releases": [
        {
          "tag": "v0.7.2",
          "kind": "patch",
          "published_at": "2026-07-20T08:30:07Z"
        },
        {
          "tag": "v0.7.1",
          "kind": "patch",
          "published_at": "2026-07-16T08:25:41Z"
        },
        {
          "tag": "v0.7.0",
          "kind": "minor",
          "published_at": "2026-07-10T01:42:00Z"
        },
        {
          "tag": "v0.7.0a3",
          "kind": "other",
          "published_at": "2026-07-08T09:28:29Z"
        },
        {
          "tag": "v0.7.0a2",
          "kind": "other",
          "published_at": "2026-07-04T16:56:35Z"
        },
        {
          "tag": "v0.7.0a1",
          "kind": "other",
          "published_at": "2026-06-27T03:12:15Z"
        },
        {
          "tag": "v0.7.0a0",
          "kind": "other",
          "published_at": "2026-06-25T21:59:18Z"
        },
        {
          "tag": "v0.6.12",
          "kind": "patch",
          "published_at": "2026-06-23T13:52:35Z"
        },
        {
          "tag": "v0.6.11",
          "kind": "patch",
          "published_at": "2026-06-21T14:19:24Z"
        },
        {
          "tag": "v0.6.10",
          "kind": "patch",
          "published_at": "2026-06-19T01:13:41Z"
        },
        {
          "tag": "v0.6.9",
          "kind": "patch",
          "published_at": "2026-06-13T06:47:46Z"
        },
        {
          "tag": "v0.6.8",
          "kind": "patch",
          "published_at": "2026-06-11T11:39:48Z"
        },
        {
          "tag": "v0.6.7",
          "kind": "patch",
          "published_at": "2026-06-04T05:35:48Z"
        },
        {
          "tag": "v0.6.6",
          "kind": "patch",
          "published_at": "2026-06-03T19:58:42Z"
        },
        {
          "tag": "v0.6.5",
          "kind": "patch",
          "published_at": "2026-06-03T02:51:02Z"
        },
        {
          "tag": "v0.6.4",
          "kind": "patch",
          "published_at": "2026-05-20T20:47:16Z"
        },
        {
          "tag": "v0.6.3",
          "kind": "patch",
          "published_at": "2026-06-09T09:19:39Z"
        },
        {
          "tag": "v0.6.2",
          "kind": "patch",
          "published_at": "2026-05-15T22:44:40Z"
        },
        {
          "tag": "v0.6.1",
          "kind": "patch",
          "published_at": "2026-05-15T14:34:38Z"
        },
        {
          "tag": "v0.6.0",
          "kind": "minor",
          "published_at": "2026-05-15T12:10:39Z"
        },
        {
          "tag": "v0.5.3",
          "kind": "patch",
          "published_at": "2026-04-25T09:57:30Z"
        },
        {
          "tag": "v0.5.2",
          "kind": "patch",
          "published_at": "2026-04-21T05:30:20Z"
        },
        {
          "tag": "v0.5.1",
          "kind": "patch",
          "published_at": "2026-04-15T19:13:03Z"
        },
        {
          "tag": "v0.5.0",
          "kind": "minor",
          "published_at": "2026-04-12T10:15:58Z"
        },
        {
          "tag": "v0.4.2",
          "kind": "patch",
          "published_at": "2026-04-11T10:37:14Z"
        },
        {
          "tag": "v0.4.1",
          "kind": "patch",
          "published_at": "2026-04-10T15:30:57Z"
        },
        {
          "tag": "v0.4.0",
          "kind": "minor",
          "published_at": "2026-04-09T15:25:50Z"
        },
        {
          "tag": "v0.3.1",
          "kind": "patch",
          "published_at": "2026-04-07T15:40:45Z"
        },
        {
          "tag": "v0.3.0",
          "kind": "minor",
          "published_at": "2026-04-07T14:03:32Z"
        },
        {
          "tag": "v0.2.6",
          "kind": "patch",
          "published_at": "2026-03-29T14:56:15Z"
        },
        {
          "tag": "v0.2.5",
          "kind": "patch",
          "published_at": "2026-03-22T19:27:22Z"
        },
        {
          "tag": "v0.2.4",
          "kind": "patch",
          "published_at": "2026-03-22T18:55:59Z"
        },
        {
          "tag": "v0.2.3",
          "kind": "patch",
          "published_at": "2026-03-22T08:15:22Z"
        },
        {
          "tag": "v0.2.2",
          "kind": "patch",
          "published_at": "2026-03-22T04:31:06Z"
        },
        {
          "tag": "v0.2.1",
          "kind": "patch",
          "published_at": "2026-03-20T14:40:59Z"
        },
        {
          "tag": "v0.2.0",
          "kind": "minor",
          "published_at": "2026-03-17T20:19:21Z"
        }
      ],
      "recent_commits": [
        {
          "oid": "4b268d1a7e9d014340bead0fa4b121a462a0708a",
          "body": "…a dimensions",
          "is_bot": false,
          "headline": "style(admin): replace native prompt() with custom modal for matryoshk…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-21T14:33:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "b221002e2ed0aec4a0995ace115c5b2161ce85ff",
          "body": "Replace grayed-out LLM test options for embedding models with\nembedding-specific test types:\n- Embedding: single input (existing)\n- Batch: array of 3 texts\n- Matryoshka: prompts user for dimensions parameter\n- Multimodal (image): sends text + image using the same test\n  image as vision tests\n\nEach type routes to /v1/embeddings endpoint. Multimodal test\nshows image preview in the test modal.",
          "is_bot": false,
          "headline": "feat(admin): embedding test menu with batch, matryoshka, multimodal",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-21T14:33:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "99a7a034b39bf6bfb88ab7722b40ed270c682766",
          "body": "…menu items\n\nFixes invalid HTML: <div> cannot be a direct child of <table>.\nBulk action bar now sits before <div class=\"table-scroll\">.\nAlso adds pointer-events:none to .test-menu-item.disabled to prevent\nclicks on grayed-out test options.",
          "is_bot": false,
          "headline": "fix(admin): move bulk-bar outside table + pointer-events on disabled …",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-21T14:33:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "90109fc809a7c4b326ed9c5ea5edbe28258efca5",
          "body": "Backend:\n- toggle_model route: per-model enabled/disabled toggle\n- bulk_update_models route: batch enable/disable/delete\n- GatewayConfig._parse_models skips models with enabled: false\n- _normalize_model_entry passes enabled field to frontend\n\nUI — model table restyle:\n- Pill toggle (ON/OFF) for per-\n[…]\nified for embedding and LLM models (embedding\n  shows Test▾ with grayed-out LLM-only options)\n- Disabled models render with reduced opacity\n\ni18n: EN/ZH strings for toggle, bulk actions, ON/OFF labels",
          "is_bot": false,
          "headline": "feat(admin): model enable/disable toggle + table UI restyle",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-21T14:33:28Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "4c1dd54963ad9f5ce40356d824c4cafd72a1115c",
          "body": "Some upstreams (e.g. gpt-5.4-mini via argo) return null for token detail\nfields like cache_write_tokens. The IR type requires dict[str, int], so\npassing raw dicts through caused a 502 validation error.",
          "is_bot": false,
          "headline": "fix: filter null values from usage token details before IR validation",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-21T12:52:36Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "579b4824daf5b9db64e139a22b08ee65276322f5",
          "body": "Add URL Template and Stream URL Template fields to the admin panel:\n- Provider modal: two optional inputs after Proxy URL\n- Provider card: displays url_template/stream_url_template when set\n- Model modal: two optional inputs after flatten_system\n- Model row: shows ⛓ url badge when url_template is co\n[…]\nand put_model accept the new fields\n- get_config normalizes url_template fields for the frontend\n- i18n: English and Chinese translations added\n\nFollows up on PR #375 which added backend-only support.",
          "is_bot": false,
          "headline": "feat: add url_template admin UI for provider and model cards",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T09:17:21Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "936c79372a6d5436c3fea79701353626b3ba16e4",
          "body": "No more open_on_no_keys=True in the gateway — if you haven't\nconfigured API keys, all /v1/* endpoints return 403. Generate\nkeys in the admin panel or set server.api_keys in config.\n\nThis is a breaking change for deployments without API keys.",
          "is_bot": false,
          "headline": "security(auth): block requests when no API keys configured",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T09:15:13Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "b430fac5307e7b178f8428353da1f7092b39df4d",
          "body": "Previously, no API keys = all requests pass through silently. Now\nAuthState accepts open_on_no_keys (default False) — downstream\nprojects must explicitly opt in to open access.\n\nThe standalone gateway sets open_on_no_keys=True for backward\ncompatibility but logs a warning at startup when no keys are\nconfigured.",
          "is_bot": false,
          "headline": "security(auth): reject requests by default when no API keys configured",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T09:11:46Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "1e6f7f61a8e63707a55052a1cf96b685432115ce",
          "body": "Same fix as config.py — testing._get_gateway_config() only checked\nthe module-level _config, breaking \"Fetch Models from Provider\" and\ntest buttons when the gateway is embedded downstream.",
          "is_bot": false,
          "headline": "fix(admin): check app.gateway_config in testing routes too",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:55:42Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9a8ead7e95d7894e09a1618d2445b47aca26b174",
          "body": "_get_gateway_config() only checked the module-level _config variable,\nwhich is None when the gateway is embedded in a downstream project\n(e.g. argo-proxy). Check request.app.gateway_config first.",
          "is_bot": false,
          "headline": "fix(admin): check app.gateway_config before module global",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:49:12Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "b966166019fbc1d5f3f81de4f13ff754b0a03773",
          "body": "Plain attribute assignment works fine since ProviderInfo is not\nfrozen/slotted. Spotted by Elena in review.",
          "is_bot": true,
          "headline": "style: simplify with_url_templates — drop unnecessary object.__setattr__",
          "author_name": "milo-oaklight[bot]",
          "author_login": "milo-oaklight[bot]",
          "committed_at": "2026-07-20T08:43:07Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "73584c5e7abf8a0149da2242dcee38b07a9191cf",
          "body": "Add url_template and stream_url_template override support at two levels:\n\n1. Per-provider: set url_template/stream_url_template in the provider\n   config block to override the registry default for all models on that\n   provider.\n\n2. Per-model: set url_template/stream_url_template in a model's dict\n \n[…]\nntries\n- GatewayConfig.resolve() applies per-model overrides via\n  ProviderInfo.with_url_templates(), which creates a shallow copy\n  sharing the same KeyRing (round-robin state preserved)\n\nCloses #374",
          "is_bot": true,
          "headline": "feat: support custom upstream URL template override in provider config",
          "author_name": "milo-oaklight[bot]",
          "author_login": "milo-oaklight[bot]",
          "committed_at": "2026-07-20T08:43:07Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "fba89d0ab7cc9d8b36063843f0c30e2667a5b883",
          "body": null,
          "is_bot": false,
          "headline": "chore: bump version to 0.7.2",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:28:52Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "f891a210f88dea4810acfdadd26be77c8cc81c1e",
          "body": "Addresses review feedback on PR #378: json.dumps doesn't escape </,\nso a branding value containing </script> would break out of the\ninjected <script> tag. Replace </ with <\\/ in the serialized JSON.",
          "is_bot": false,
          "headline": "fix(admin): escape </ in branding JSON to prevent script tag breakout",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:28:33Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "8e39dc09a73c1b5ddf2a2e7b3bd415fdb56fd7e8",
          "body": "Downstream projects can pass branding={title, subtitle, version,\nlinks, attribution} to setup_admin(). The dict is serialized as\nwindow.__branding in custom_head, and a consumer script in admin.html\npatches the header, login screen, and settings footer.",
          "is_bot": false,
          "headline": "feat(admin): add branding dict for downstream identity customization",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:14:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "6152f22ee5bbb397d68b87bf852a56e6910c5738",
          "body": "The magnifying glass view button used a Unicode escape (🔍)\nwhich renders inconsistently across platforms. Replace with an inline\nSVG that matches the rest of the admin panel's icon style.",
          "is_bot": false,
          "headline": "style(admin): replace emoji icon with inline SVG in capture table",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:14:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "ebf01953be8a915ff6e8c6b4cbb1c3a4c8a2c66d",
          "body": "setup_admin() now accepts an optional custom_head parameter — an HTML\nfragment (style/script tags) injected before </head> when serving the\nadmin panel. Downstream projects can override UI behavior without\nmodifying the reference admin.html.\n\nThe HTML is cached per custom_head value, so there's no per-request\noverhead.",
          "is_bot": false,
          "headline": "feat(admin): add custom_head injection for downstream UI overrides",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T08:14:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "787bcf30de9b7055cdf403e1892fdb6a1150446c",
          "body": "Mirrors the _get_config_io pattern: raise with a descriptive message\ninstead of returning None. Removes 16 redundant None-guard blocks\nacross route handlers. The observability fallback wraps the call in\ntry/except to preserve its soft-failure semantics.",
          "is_bot": false,
          "headline": "refactor: _get_config_path raises RuntimeError on missing path",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T05:04:23Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "edfe010bdcf091f8933d6bf03bd409b19c90264a",
          "body": null,
          "is_bot": false,
          "headline": "fix: add None guard to _get_config_io with descriptive RuntimeError",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T05:04:23Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "8a4be00710c63d4137df733ffb47aeecac450977",
          "body": "Admin route handlers now use app.config_io (a ConfigIO protocol\ninstance) instead of importing load_config_raw/write_config directly.\nThis allows downstream projects like argo-proxy to supply their own\nI/O adapter for non-JSONC config formats (YAML, TOML, etc.).\n\n- ConfigIO protocol: load(), load_ra\n[…]\nimplementation (backward compatible)\n- setup_admin() accepts optional config_io parameter\n- All admin routes (config, keys, auth, observability) use\n  _get_config_io(request) instead of direct imports",
          "is_bot": false,
          "headline": "refactor: extract ConfigIO protocol for admin panel config I/O",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-20T05:04:23Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "67bd912f7cac3169cdd1a1870a533734acc51728",
          "body": "Fixes fd leak: close() now closes _async_writer so __del__\ncan clean up leaked async streaming responses.",
          "is_bot": false,
          "headline": "chore(vendor): update httpclient 0.4.4 → 0.4.5",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-19T11:35:10Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "8e9184da41733b50e64ff5d915030c2064580461",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.1",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T08:24:21Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "a954be66dd6947ace05980252ef6e0741af99ea9",
          "body": "When nullable: true appears on a schema with anyOf/oneOf but no type\nfield, inject {\"type\": \"null\"} into the variants list instead of\nsilently dropping the nullability information.\n\nAddresses Clementine's review feedback on PR #373.",
          "is_bot": false,
          "headline": "fix: handle nullable + anyOf/oneOf without type field",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T08:17:52Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "f29a0ff4f6c125eabf4b2f81513091cedb88ad72",
          "body": "…(#372)\n\nAnthropic rejects the OpenAPI `nullable` extension in tool parameter\nschemas. Add `convert_nullable_to_type_array()` helper that recursively\nconverts `\"nullable\": true` to standard JSON Schema `\"type\": [T, \"null\"]`.\n\n- Anthropic converter: strip `title`, convert `nullable` to type arrays\n- Google GenAI converter: strip `title` (nullable is kept — Google supports it)\n- OpenAI converters: no changes needed (lenient validation)",
          "is_bot": false,
          "headline": "fix: strip nullable and title from tool schemas for Anthropic/Google …",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T08:17:52Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "27dfb23e7ad5195d6e265f20411efab3cd4721a5",
          "body": "- scripts/scan_sdk_types.py: introspects upstream SDK types vs our\n  TypedDict replicas, reports field drift and new relevant types\n- .github/workflows/sdk-type-scan.yml: manual-trigger workflow that\n  runs the scan and posts the report to the workflow summary",
          "is_bot": false,
          "headline": "feat: add SDK type coverage scanner and manual CI workflow",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:57:23Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "6af66017062694ec50eaea4e8ea21f14505db29b",
          "body": "SDK compatibility is enforced by weekly sdk-compatibility CI, not version pins.",
          "is_bot": false,
          "headline": "revert: drop LLM SDK version floors",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:35:03Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "d6b03df32a1dcb9a0d4827be24343b24fa253c64",
          "body": "SDK compatibility is enforced by the weekly sdk-compatibility CI job,\nnot by version constraints. Pinning floors adds maintenance burden\nwith no real benefit.",
          "is_bot": false,
          "headline": "revert: drop LLM SDK version floors — unpinned is latest by default",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:34:53Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "f4b3280b72642ea140277f0bb17d117c755ef0b8",
          "body": "Tested with openai 2.45.0, anthropic 0.116.0, google-genai 2.12.0 —\nall 2108 tests pass. Set floor versions to the last known-good baseline.",
          "is_bot": false,
          "headline": "chore: add minimum version constraints for LLM SDK extras",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:33:24Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "180b476fd934aee8e22de633a7e3e47851c2681a",
          "body": "- Add allow list to pip ecosystem (openai, anthropic, google-genai)\n- Remove github-actions ecosystem (noisy, low-value updates)",
          "is_bot": false,
          "headline": "chore: restrict dependabot to LLM SDK deps only",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:28:36Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "c7c21cf974c768528ef495f35f9ce1a05ceee6b6",
          "body": null,
          "is_bot": false,
          "headline": "chore: bump vendored validate 0.6.0 → 0.6.1 (dataclass instance support)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-16T06:12:30Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "81ed15999e9a9c40082a7392bd0c0366c18960a0",
          "body": "- Use checkbox-group pattern matching capabilities section\n- Add \"System Options\" section label\n- Add Chinese i18n for systemOptions, flattenSystem, flattenSystemHint",
          "is_bot": false,
          "headline": "fix: flatten_system checkbox layout and i18n",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T05:43:13Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "c52c833a1165f3950c3f1f6317e6d7debd4ca823",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.0",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T01:40:32Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "f8f6e618dd3c69edac666bbb8d9690838c096fe4",
          "body": "- Always send flatten_system boolean (not only when checked), so\n  unchecking can override gemini auto-detection\n- Populate checkbox when editing existing models (read from config,\n  fallback to gemini auto-detect)",
          "is_bot": false,
          "headline": "fix: admin panel flatten_system checkbox bugs",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T01:32:54Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "b9a27082550ae010400f681d4941f29b40522e4a",
          "body": "- OpenAI Chat converter now outputs structured content arrays for system\n  messages (preserving block boundaries for cache_hint passthrough)\n- New flatten_system_content() body-level transform factory to flatten\n  system content arrays to plain strings when needed\n- Per-model flatten_system config f\n[…]\nth auto-detection\n  for gemini models\n- Admin panel checkbox toggle for flatten_system\n- Argo shim: flatten_system_content(pattern=r\"^gemini\") as default\n  post_ir_transform for Gemini models via Argo",
          "is_bot": false,
          "headline": "feat: flatten_system_content transform + per-model gateway toggle (#370)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T01:32:54Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "6da24a0d8b44078d553dd25db4fe581b650554de",
          "body": null,
          "is_bot": false,
          "headline": "fix: add proper type annotations in test_cache_hint for ty check",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T00:09:39Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "405f7810251df39de7a0a5f9b3383a24c1a4c500",
          "body": "Add cache_hint field to IR types and bidirectional cache_control ↔\ncache_hint mapping in the Anthropic converter. This preserves\nAnthropic's block-level prompt caching markers through IR round-trips.\n\nIR types — added cache_hint: NotRequired[dict[str, Any]] to:\n  TextPart, ImagePart, FilePart, ToolC\n[…]\nmethods don't read it.\n\n37 new tests covering block-level round-trip, full request round-trip,\nselective preservation, and cross-format harmlessness.\n\nCloses #362. Downstream: Oaklight/argo-proxy#140.",
          "is_bot": false,
          "headline": "feat: preserve Anthropic cache_control through IR via cache_hint (#362)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-10T00:09:39Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "41693e24ab8723a72e2b52a35b6b7b787fad4387",
          "body": "Update test_shim_loader.py, test_transforms.py, test_shims.py, and\ntest_proxy_transforms.py to use pre_ir_transforms/post_ir_transforms\nconsistently instead of the deprecated aliases. Fixture strings in\ntest_shim_loader.py also updated so they no longer trigger deprecation\nlogs during test runs.",
          "is_bot": false,
          "headline": "chore: migrate remaining tests to new transform names",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:48:03Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "06a66afee2b6037293afdbf4d49ce73ed2634948",
          "body": "…w names\n\n- Add DeprecationWarning when ProviderShim is constructed with legacy\n  from_transforms/to_transforms kwargs\n- Add logger.info deprecation notice when transforms.py exports\n  legacy names\n- Rename all 11 built-in provider transforms.py to use\n  pre_ir_transforms/post_ir_transforms\n- Update test constructors and method names to use new names",
          "is_bot": false,
          "headline": "chore: add deprecation warnings and migrate provider transforms to ne…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:48:03Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "0fe65a2c9c31dcd15adef6431b48a5bb1b5c4a96",
          "body": "- Fix _load_transforms fallback: use `is None` instead of `or` to\n  correctly handle empty tuple () as a valid explicit value (not\n  fallthrough to legacy name)\n- Reject unknown kwargs in ProviderShim.__init__ to match frozen\n  dataclass behavior\n- Add \"Keep in sync\" comment on defaults dict\n\nAddresses review feedback from Elena and Milo on PR #368.",
          "is_bot": false,
          "headline": "fix: use is None check in _load_transforms and reject unknown kwargs",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:48:03Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "5af5140c7d8583284c608062f32a1fab87838e76",
          "body": null,
          "is_bot": false,
          "headline": "test: add cache_write_tokens coverage to chat completions type tests",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:10:32Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "29f44d81e9149bf51d001db1e13cf740233e383c",
          "body": "Rename body-level transform fields to clarify their position in the\nconversion pipeline:\n\n  from_transforms → pre_ir_transforms  (applied BEFORE IR conversion)\n  to_transforms   → post_ir_transforms (applied AFTER IR conversion)\n\nThe old names were ambiguous — \"from\" and \"to\" described data flow\ndir\n[…]\nes remain\n  as read-only aliases\n- Provider transforms.py files can export either name (new takes\n  precedence)\n\nAll internal usage (pipeline.py, auto_detect.py, provider loader)\nupdated to new names.",
          "is_bot": false,
          "headline": "refactor: rename from/to_transforms to pre/post_ir_transforms",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:08:35Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "1eccf1245ce2dba134a096046ec4e98a9f98ff36",
          "body": "OpenAI SDK >=2.35 added cache_write_tokens as a required field in\nResponses InputTokensDetails and optional in Chat PromptTokensDetails.\nUpdate TypedDict replicas and test fixtures accordingly.",
          "is_bot": false,
          "headline": "fix: add cache_write_tokens field for OpenAI SDK 2.45+ compat",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T23:08:16Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "ec087e1872e738d0f8ab1c004409dc4b978637ad",
          "body": "- Fix ruff format on test_ir_types.py (line length)\n- Fix ty invalid-argument-type errors in OpenAI chat type\n  compatibility tests: use SDK types for annotation to avoid\n  cross-type unpacking issues when SDK content types evolve\n  (openai 2.44.0 → 2.45.0 changed content field types).\n  Rosetta type existence is still verified via __annotations__.",
          "is_bot": true,
          "headline": "fix: ruff format + ty compat with openai SDK 2.45.0",
          "author_name": "milo-oaklight[bot]",
          "author_login": "milo-oaklight[bot]",
          "committed_at": "2026-07-09T23:08:16Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "547daec51d80ff559dfece8ba9237102804c1296",
          "body": "Two test files still used plain string for system_instruction in\ntyped IRRequest literals, causing ty invalid-argument-type errors\nafter #365 changed the canonical form to list[TextPart].",
          "is_bot": true,
          "headline": "fix: update test system_instruction to list[TextPart] for ty check",
          "author_name": "milo-oaklight[bot]",
          "author_login": "milo-oaklight[bot]",
          "committed_at": "2026-07-09T23:08:16Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "0a70ba4004c9f7c234e99964da37f159f983e54c",
          "body": "OpenAI Chat system messages should use plain string content for\nmaximum downstream compatibility — some consumers (e.g. argo-proxy\nrouting to Gemini) don't handle the array content form.",
          "is_bot": false,
          "headline": "fix: output system as plain string in OpenAI Chat converter",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T06:02:35Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "5fc015c4d667a9028fbebf4a3d55d02bf242524c",
          "body": "Change the canonical IR form of system_instruction from str to\nlist[TextPart], following the principle established in #69 that IR\nshould have one canonical form.\n\nA single string \"You are helpful.\" is now represented as:\n  [TextPart(type=\"text\", text=\"You are helpful.\")]\n\nThis ensures consistent str\n[…]\nResponses: produces list[TextPart]; flattens to string for\n  instructions field\n- Google GenAI: _parse_system_instruction returns list[TextPart];\n  request_to_provider builds parts array\n\nCloses #364.",
          "is_bot": false,
          "headline": "refactor: unify system_instruction to list[TextPart] (#364)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-09T06:02:35Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "d0e556589c41a563e636a57d67ae8f5c67e37986",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.0a3",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-08T09:27:25Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "4329cb085539a1e04853e34f7a5d6b7127d4f266",
          "body": "Both error return paths in handle_non_streaming (UpstreamConnectionError\nand upstream 4xx/5xx) now record capture data. Error responses are often\nthe most valuable for debugging/inspection.",
          "is_bot": true,
          "headline": "fix: capture error responses in non-streaming path",
          "author_name": "clementine-oaklight[bot]",
          "author_login": "clementine-oaklight[bot]",
          "committed_at": "2026-07-08T08:37:04Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "fca80295af1104a635e844680f9f91c08485835c",
          "body": "status() now acquires the lock for external callers.\nenable()/disable() call _status_unlocked() internally since\nthey already hold the lock. threading.Lock is not reentrant,\nso the previous pattern would have deadlocked if status()\never tried to acquire the lock.",
          "is_bot": true,
          "headline": "fix: split CaptureState.status() to avoid Lock reentrancy",
          "author_name": "clementine-oaklight[bot]",
          "author_login": "clementine-oaklight[bot]",
          "committed_at": "2026-07-08T08:37:04Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9915efae4a898177e4bff312cc5d4c1ff5ccf058",
          "body": "Add CaptureState (same pattern as ProfilerState) for capturing the\nfull three-stage request flow in memory:\n- original_request (client body as received)\n- converted_body (target-format after pipeline conversion)\n- upstream_response (provider response body)\n\nEnable for N requests via admin API, view \n[…]\ntureState on app\n- admin/routes/__init__.py — register capture API endpoints\n- admin.html — Dashboard UI section with enable/disable toggle,\n  results table, and collapsible detail viewer\n\nCloses #359",
          "is_bot": true,
          "headline": "feat: in-memory content capture for request/response inspection",
          "author_name": "clementine-oaklight[bot]",
          "author_login": "clementine-oaklight[bot]",
          "committed_at": "2026-07-08T08:37:04Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "d8447a835d8d48c787ba445b7f05a58b226c82ec",
          "body": "* feat: add error detail log toggle to settings popup\n\nNew toggle in Settings → \"Error Detail Log\" (default: enabled).\nWhen disabled, error dumps are not written to SQLite on upstream\nfailures — reduces storage for high-error-rate deployments.\n\n- GatewayConfig: debug.error_dumps (default true)\n- app\n[…]\nssing unrelated persistence\nwrites if the proxy path ever gains additional uses for the object.\n\n---------\n\nCo-authored-by: elena-oaklight[bot] <294684326+elena-oaklight[bot]@users.noreply.github.com>",
          "is_bot": false,
          "headline": "feat: add error detail log toggle to settings popup (#358)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-08T07:38:52Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "0c452a5a9422ed1b47d790f788f5b64f609a5183",
          "body": "Settings popup expanded from 2 to 7 operational controls:\n- Admin password change (current password required, rate-limited)\n- Admin token copy + force-rotate + auto-rotate countdown\n  (configurable 5/15/30/60 min, monospace digits with CJK text)\n- Log level selector (Normal / Verbose / Debug with bo\n[…]\nss_max/error_max setters\n\nCLI:\n- llm-rosetta-gateway set-password — forgot-password recovery\n  (no current password, interactive getpass, min 4 chars)\n\nFull bilingual i18n (en/zh) for all new strings.",
          "is_bot": false,
          "headline": "feat: enhanced settings popup — password, token, log level, retention",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-07T23:29:40Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "139144e9544cc07bfe3b159c379cf8df9d1b521c",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.0a2",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-04T16:55:37Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "446fcc27aab10fa874f218bfdb939eac97826853",
          "body": null,
          "is_bot": false,
          "headline": "feat: add stream_connect_ms to streaming profile",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-04T16:35:18Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "683c92ef7c07be9ecbe3b11ba7ed17d72ad4a752",
          "body": "Picks up CaseInsensitiveDict for response headers and\ncase-insensitive header dedup (zerodep #116, #117).",
          "is_bot": false,
          "headline": "chore: update vendored httpclient 0.4.2 → 0.4.4",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-04T06:00:17Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "281586a6ab159ba624e0e2cd32afba32bd9f1247",
          "body": "Re-vendors zerodep validate with the discriminated union dispatch\ntable optimization. _try_discriminated now uses a cached\n{literal_value: TypedDict} lookup instead of O(variants) linear scan.\n\nProduction profiling showed 93% of conversion time (1079ms/1165ms)\nspent in _validate_union on large agent conversations. This fix\nreduces that to ~13ms for 1000-message payloads.\n\nzerodep#107, zerodep#108",
          "is_bot": false,
          "headline": "vendor: update validate 0.5.0 → 0.6.0 (O(1) union dispatch)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-07-02T02:55:03Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "54f807f3cfeec48242d544d67ddc282484892e6f",
          "body": "… proxy\n\nAdd comments explaining why UpstreamConnectionError synthesizes an error\nmessage (no upstream response exists) while stream.is_error passes the\nupstream body through as-is (transparent proxy for client SDK parsing).",
          "is_bot": false,
          "headline": "docs: clarify connection-level vs application-level error handling in…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-29T08:04:42Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9a0ce8f38295dc527d507f65245c2106cd102832",
          "body": "Move upstream connection opening from _stream_event_generator into\nhandle_streaming so errors (4xx/5xx) are detected before the HTTP 200\nStreamingResponse headers are sent. Previously, upstream errors like\ntoken-limit 400s were buried inside SSE data events on an HTTP 200\nresponse, causing client SDKs to report misleading \"Invalid JSON body\"\ninstead of the actual error message.\n\nCloses #345",
          "is_bot": false,
          "headline": "fix: return correct HTTP status for streaming upstream errors",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-29T03:00:19Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "47e6960fac1099072cfb2acbe41120e81e89961d",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.0a1",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-27T03:10:56Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "02a7c2e7a36bb13a8f12eda7483d5e354630d23d",
          "body": "zerodep/zerodep#105: _SSEParser now accepts last_id and retry as\nconstructor kwargs instead of writing to private slots after init.",
          "is_bot": false,
          "headline": "fix: update vendored sse to 0.3.2 — use constructor args for parser init",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9c15e5a9e56e1a6829cc328a7cdb668f0d8f528d",
          "body": "… of silent disconnects\n\nzerodep/zerodep#106: TimeoutError, IncompleteReadError, and uncaught\nexceptions during request parsing now return 408/400/500 JSON responses\ninstead of silently closing TCP. Fixes \"Connection error\" cascades when\nclients (Claude Code) go through NPS tunnels.",
          "is_bot": false,
          "headline": "fix: update vendored httpserver to 0.2.1 — return HTTP errors instead…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "c824f71c3b7f8ae000360b4a9db7daceb4360cac",
          "body": "Add assert-not-None guards before subscripting get_error_dump() and\nget_dump_body() results, and assert isinstance for body_hash before\npassing to get_dump_body(). Fixes 10 ty type-check diagnostics.",
          "is_bot": false,
          "headline": "fix: add None guards in test_error_dump.py for ty check",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "df7ec33da61fc9921296f544bc5b39221b5b169e",
          "body": "- Remove private symbols (_Bucket, _RollingWindow, _ProviderStats) from\n  gateway/admin/metrics.py __all__ — they remain importable via noqa\n  but aren't part of the public surface\n- Replace incorrect '.. deprecated::' Sphinx directives with plain\n  docstring notes in re-export shims (old paths are not deprecated,\n  they've been published since v0.5.0)\n- Rewrite test_max_results_trim to exercise store_result() with a\n  fake profiler instead of manually appending and trimming",
          "is_bot": false,
          "headline": "fix: address review feedback on observability extraction",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "04a4850b27e0a62a71300cf3956b53e821a68126",
          "body": "Store full request context (body, upstream response, metadata) when\nproxy requests fail, enabling replay and post-mortem debugging.\n\nDesign decisions:\n- SQLite storage in existing gateway.db (dump_bodies + error_dumps)\n- zlib level 6 compression for request/converted bodies\n- Base64 image offload (r\n[…]\nwnload raw request body JSON\n  DELETE /admin/api/error-dumps           — clear all\n\nIncludes 25 tests covering offload, hashing, compression, dedup,\nfire-and-forget safety, and persistence operations.",
          "is_bot": false,
          "headline": "feat: add error dump capability for failed proxy requests",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "0e737e7a8ae375ec97d00ae6f99144fde9267a76",
          "body": "Move MetricsCollector, RequestLog, RequestLogEntry, PersistenceManager,\nand ProfilerState (data layer) from gateway/admin/ to a new top-level\nllm_rosetta/observability/ package.\n\nThese modules have zero coupling to GatewayConfig or the httpserver\nframework, making them directly reusable by external \n[…]\n from observability\n- gateway/app.py → imports RequestLogEntry from observability\n- NEW: tests/observability/ with 34 tests covering standalone usage\n  and backward-compat identity checks\n\nCloses #341",
          "is_bot": false,
          "headline": "refactor: extract reusable observability layer from gateway admin",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T23:10:43Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "f16d9b03036a01c2952f7244b6a6aab2372e1155",
          "body": "- Remove invalid type annotation on subscripted expression (invalid-type-form)\n- Use explicitly typed local variable for adaptive thinking dict to avoid\n  dict[str, str] inference causing invalid-assignment downstream\n- Relax test mock return type to list[Any] since it returns plain dicts",
          "is_bot": false,
          "headline": "fix: resolve ty 0.0.54 type-check errors after dev tool version pin",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T22:07:47Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "36b63a8c5b68f181eb7250fca34c3e9e10850144",
          "body": null,
          "is_bot": false,
          "headline": "pin dev tool versions: ruff==0.15.20, ty==0.0.54",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T21:19:21Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "e07f35386edc709bd07b8b7559d3ec5b585c6e73",
          "body": "Use fcntl.flock(LOCK_EX) to prevent concurrent writes when multiple\ngateway instances share the same config file via hard link. Opens with\n\"a+\" to avoid truncating before the lock is acquired.",
          "is_bot": false,
          "headline": "fix: add file lock to write_config for cross-process safety",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T17:42:42Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "0712622195187f08bcc7ac4dde298de37743142a",
          "body": "Button next to Per-Provider Breakdown title calls POST /admin/api/metrics/rebuild.\nShows toast with row count on success. i18n: EN + ZH.",
          "is_bot": false,
          "headline": "feat: add Rebuild Counters button to Dashboard breakdown section",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T16:48:26Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "f3ebb4af21e143074eef02c6a61b05cdd166dd22",
          "body": "- iter_log_rows_for_rebuild: fetchmany(5000) batched iteration instead\n  of unbounded fetchall — bounded memory regardless of table size\n- rebuild_counters: build in temporaries then swap atomically, so\n  concurrent snapshot() calls never see half-rebuilt state\n- rebuild_counters accepts Iterable (generators work), returns count\n- active_streams preserved (live state, not historical — commented)",
          "is_bot": false,
          "headline": "fix: rebuild_counters uses streaming iteration and atomic swap",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T16:33:59Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "e2dd76234eb70c14862b3757415ecda787e442b0",
          "body": "by_target_provider was keyed by API type (e.g. \"anthropic\") instead of\nprovider name (e.g. \"Argo Claude\"), causing all providers sharing the\nsame API type to be merged in the dashboard breakdown.\n\nAlso adds POST /admin/api/metrics/rebuild to reconstruct counters from\nrequest log entries — useful aft\n[…]\nilable)\n- MetricsCollector.rebuild_counters: recompute all counters from log rows\n- PersistenceManager.all_log_rows_for_rebuild: lightweight query\n- New admin endpoint: POST /admin/api/metrics/rebuild",
          "is_bot": false,
          "headline": "fix: metrics breakdown uses provider name instead of API type",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T16:33:59Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "17136776e4f4c55201db37f2572fa206691ed267",
          "body": "- Single download button per result (SVG download icon)\n- Download All button (ZIP archive via server-side endpoint)\n- New endpoint: GET /admin/api/profiling/results/download (ZIP)\n- Hint text: \"Results are in-memory only, lost on restart\"\n- Download All button auto-shown/hidden based on results count\n- i18n: all new keys in both EN and ZH",
          "is_bot": false,
          "headline": "feat: profiling UI — download single/all flamegraphs, restart hint",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T08:45:34Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "ea79659ecfa50876bd93256665c4844801255ca4",
          "body": "- viewFlamegraph: fetch with x-admin-token header via _adminHeaders(),\n  write HTML into new window (previously window.open sent a bare GET\n  without auth, causing 401)\n- Replace fire emoji with SVG flame icon on View button\n- Remove mode column emojis (⇅/⬡)\n- toggleProfiling: show toast on enable error (e.g. pyinstrument missing)",
          "is_bot": false,
          "headline": "fix: flamegraph viewer sends admin auth, replace emoji with SVG",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T08:37:51Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "2d9474fd45095789016921233560ce75d7b37420",
          "body": "- Docker: install [gateway,profiling] instead of just [gateway]\n- enable_profiling: pre-check pyinstrument availability, return 400\n  with install instructions if missing (not silent failure)\n- _try_start_profiler: restore consumed slot on RuntimeError so the\n  counter doesn't decrement without producing a result",
          "is_bot": false,
          "headline": "fix: Docker image includes pyinstrument, enable endpoint pre-checks",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T08:32:06Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "79b886867a975c0e436dee21c908418185751dd6",
          "body": "Add interactive profiling controls to the Dashboard tab:\n- Enable/disable button with request count input\n- Status badge showing ON/remaining or OFF\n- Results table with timestamp, model, route, mode, duration\n- Flamegraph button opens pyinstrument HTML in new tab\n- Clear button to reset results\n- Auto-refreshes with dashboard metrics (3s interval)\n- i18n: English + Chinese translations",
          "is_bot": false,
          "headline": "feat: add Profiling section to admin Dashboard UI",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T08:21:34Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "3e987024e50c4e155ceb6622a19616bb46f6a260",
          "body": "- Use request.path_params[\"index\"] directly instead of double-lookup\n  fallback through kwargs\n- Use _qp() helper for query param access instead of custom hasattr\n  check, matching other admin route handlers",
          "is_bot": false,
          "headline": "fix: align profiling handler with codebase conventions",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T08:12:06Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "e09e091d6d794ea0bb05c11e5297590d634a4943",
          "body": "…trument\n\nTwo-layer profiling system for diagnosing conversion and upstream latency:\n\nCore library:\n- ConversionPipeline.profile: always-on perf_counter timing for each\n  conversion sub-phase (source_to_ir, ir_transforms, ir_to_target, etc.)\n- DeepProfiler: pyinstrument wrapper with sync/async conte\n[…]\nadmin/api/profiling/status\n  GET  /admin/api/profiling/results\n  GET  /admin/api/profiling/results/<index>\n  DELETE /admin/api/profiling/results\n\npyinstrument is a new [profiling] optional dependency.",
          "is_bot": false,
          "headline": "feat: add hybrid profiling — always-on phase timing + on-demand pyins…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-26T07:54:46Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "cfac5dd11f3233db52599044ccd2b35878ce9224",
          "body": "Integration tests depend on external packages (agentabi) not available\nin CI. The llm_api_simple_tests submodule has its own code style.\nAlso apply ruff format fixes to test_gateway_agentabi.py.",
          "is_bot": false,
          "headline": "fix(ci): exclude integration tests and submodule from ruff/ty checks",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T23:22:12Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "7e0c00e32d30baf7bdafef74b93b3c15ee5cb28e",
          "body": "- embeddings.py: migrate from resolve_model() to resolve() API,\n  replace inline telemetry with shared _record_telemetry()\n- config.py: remove resolve_model() (no remaining callers),\n  remove duplicate DEFAULT_CAPABILITIES class variable",
          "is_bot": false,
          "headline": "refactor: clean up post-#320 gateway remnants",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T22:53:13Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "6772df6a7e9d52114d8f588994d1552f32e96689",
          "body": "Uniform model coverage across codex, opencode, and claude_code agents.\nRequires ROSETTA_DEV_TEST_BASE_URL and ROSETTA_DEV_TEST_KEY env vars.\nNot run by default (tests/integration/ is excluded from make test).",
          "is_bot": false,
          "headline": "test: add agentabi E2E gateway test matrix",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T22:04:36Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "42cc557cc70d79622eda0cd9622268989af16c46",
          "body": null,
          "is_bot": false,
          "headline": "release: v0.7.0a0",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T21:58:20Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "1b46918e07f75bf9c878ada0dc2e2b4e7be321bc",
          "body": "OpenAI Responses API rejects reasoning.type with 400 Unknown parameter.\nReasoning is controlled via reasoning.effort only. Remove the type\nfield emission from _apply_openai_responses_extras.",
          "is_bot": false,
          "headline": "fix: stop emitting reasoning.type for OpenAI Responses API",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T15:01:27Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "e7393224c5daada742250de10cae74729782771d",
          "body": "Google GenAI API rejects JSON Schema draft 6+ numeric constraints\n(exclusiveMinimum, exclusiveMaximum) in tool definitions. Add them\nto UNSUPPORTED_SCHEMA_KEYS so sanitize_schema strips them before\nforwarding to upstream providers.",
          "is_bot": false,
          "headline": "fix: strip exclusiveMinimum/exclusiveMaximum in sanitize_schema",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T15:01:27Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "62966931ced78191cbb1a6fd0d5aad12bc721943",
          "body": "… (#335)\n\nIntroduce capabilities.py as a platform-level capability enforcement\nmodule, distinct from shim transforms (provider dialect adaptation).\n\n- enforce_reasoning(ctx, shim): configure reasoning output mode (pre-IR)\n- enforce_vision(ir, model_capabilities): strip images for non-vision\n  models\n[…]\n deprecated alias for enforce_reasoning\n\nRemoved strip_non_vision_images() from argo shim ir_transforms — it's\nnow platform-level behavior handled by enforce_vision, not a\nprovider-specific transform.",
          "is_bot": false,
          "headline": "refactor: extract capabilities.py — enforce_reasoning, enforce_vision…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-25T08:33:47Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "fe6b93682c1e824bb91ae653d641666e21ba2d6c",
          "body": "Naming:\n- apply_shim_to_ir → apply_ir_transforms (drop misleading shim prefix)\n- setup_shim_context → configure_context (pipeline configures, not shim)\n- Old names kept as compat aliases with DeprecationWarning\n\nNew IR transform system (separate from body-level Transform):\n- TransformContext: carrie\n[…]\nlts empty)\n\nDesign: body transforms (1-arg) and IR transforms (2-arg) are separate\ntypes with separate executors — no dispatch magic, no signature\ninspection, no breakage of existing custom callables.",
          "is_bot": false,
          "headline": "refactor: rename pipeline API + add IRTransform system (#330)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T23:50:13Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "bc3cd3e1489d202c86b758f9f899c8f1533f8f5c",
          "body": null,
          "is_bot": false,
          "headline": "docs: fix default_message_field docstring and test name per review",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T20:42:47Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "18eebc5bc778f5a08452ab12ab197b673847a7ef",
          "body": "…ves (#328)\n\nNew Transform factory functions in shims/transforms.py:\n\n- replace_message_field(field, old, new): replace field value in all\n  messages[] entries (e.g. \"developer\" → \"system\" role downgrade)\n- default_message_field(field, default): set default for messages[].field\n  when value is None \n[…]\no_transforms, making\npreviously hardcoded argo-proxy body fixups (_downgrade_developer_role,\n_normalize_null_content, _strip_temperature_for_reasoning_models)\navailable declaratively through the shim.",
          "is_bot": false,
          "headline": "feat(shim): add message-level and model-conditional transform primiti…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T20:42:47Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "5cc7a090972350001a16efcfc97b232d02fd7e0e",
          "body": "…eview\"\n\nThis reverts commit 08485e7bcffdbcaca81e35e2b80d59f827501785.",
          "is_bot": false,
          "headline": "Revert \"docs: fix default_message_field docstring and test name per r…",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T19:51:47Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "08485e7bcffdbcaca81e35e2b80d59f827501785",
          "body": null,
          "is_bot": false,
          "headline": "docs: fix default_message_field docstring and test name per review",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T19:50:42Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "d1a7d6e6c62cc58dda942188d6ea97e9cf7a3472",
          "body": null,
          "is_bot": false,
          "headline": "fix(ty): remove unused type-ignore, type-annotate test helpers",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T18:31:23Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "a35cedd6f04e04ea5231bfa788a4f6ab5f800427",
          "body": "Rebased on top of #322 (ConversionPipeline). Introduces:\n\n- routing.py: ResolvedRoute (frozen dataclass) + Router (Protocol)\n  in core library — zero network deps\n- GatewayConfig.resolve(source, model) → (ResolvedRoute, ProviderInfo)\n  consolidates model lookup + shim + capabilities + reasoning over\n[…]\nnly\n(protocol contract); GatewayConfig.resolve() returns the tuple as a\nconcrete convenience — docstring clarifies the distinction.\n\nBackward compat: resolve_model() retained for embeddings and admin.",
          "is_bot": false,
          "headline": "refactor: extract routing layer — ResolvedRoute + Router protocol (#323)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T18:31:23Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "929261dc1c4da986d4a299a41d213499b412bf8f",
          "body": null,
          "is_bot": false,
          "headline": "fix: guard convert_request() against double-call (one-shot pipeline)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T11:55:15Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9a93cb77d356a2277aa923bbd5909c066b40f81d",
          "body": "Add ConversionPipeline, StreamProcessor, and ConversionError to the\ncore library's pipeline.py module. This encapsulates the full\nPhase 1→2→4 conversion lifecycle:\n\n- convert_request(): source body → target body (Phase 1+2)\n- convert_response(): target response → source response (Phase 4)\n- create_s\n[…]\nublished, pipeline is top-level)\n- Remove pipeline re-exports from shims/__init__.py\n- Rename tests/test_shim_pipeline.py → tests/test_pipeline.py\n- Add 14 new ConversionPipeline/StreamProcessor tests",
          "is_bot": false,
          "headline": "refactor: extract ConversionPipeline class into pipeline.py (#322)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T11:55:15Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "6001cef4c97b716ef190968b5818dee050cdd61d",
          "body": "Address review feedback from #329:\n\n1. embeddings.py: replace direct HttpTransport._pool access with\n   transport.send_passthrough() — a new UpstreamTransport method for\n   non-conversion passthrough endpoints (embeddings, future reranking).\n   The handler is now transport-agnostic.\n\n2. Rename _openai_auth/_anthropic_auth/_google_auth to public names\n   (drop underscore). These are consumed by providers.py's registry\n   and should be part of the transport module's public API.",
          "is_bot": false,
          "headline": "fix: embeddings uses transport abstraction; make auth functions public",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T10:45:28Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "64a9ae8763dceb866c9546d315eccbad441aa026",
          "body": "Introduce a proper transport layer with protocol abstraction, HTTP/SSE\nimplementation, and vendored W3C SSE parser.\n\nArchitecture:\n- transport/_base.py: UpstreamTransport Protocol, UpstreamResponse,\n  UpstreamStream ABC — proxy.py programs against this interface\n- transport/http/: HttpTransport impl\n[…]\ndown\n- prepare_upstream/get_client/SSE parsing absorbed into HttpTransport\n\nFuture protocol backends (gRPC, WebSocket) can be added as sibling\npackages under transport/ implementing UpstreamTransport.",
          "is_bot": false,
          "headline": "refactor(gateway): extract layered transport abstraction (#321)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T10:45:28Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "15863dd738f63a8380d8a0eab4fd0885b7af755c",
          "body": null,
          "is_bot": false,
          "headline": "docs: clarify private helper imports in test module docstring",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T10:01:18Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "2b536affe6b41aa4551777c9f641da8cbbb2c927",
          "body": "Pipeline is orchestration logic that consumes Shim config — it belongs\nas a peer to shims/ and converters/, not inside the shims/ package.\n\n- Move shims/pipeline.py → pipeline.py (top-level module)\n- Old path (shims/pipeline.py) becomes backward-compat re-export\n- Fix apply_shim_to_ir docstring: \"always use the return value\"\n- Update gateway/proxy.py and tests to import from new location",
          "is_bot": false,
          "headline": "refactor: move pipeline.py to top-level, fix docstring (#319)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T10:01:18Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "409c9c8639e211a1d6209a88ce55e4bca3455a33",
          "body": null,
          "is_bot": false,
          "headline": "fix(ty): add type annotations to test helpers for ty check",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T07:38:35Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "9d52a43ade2efdc6648815b84c4548b11461e20b",
          "body": "Create two public entry points in `shims/pipeline.py`:\n\n- `setup_shim_context(ctx, shim)` — inject reasoning config into\n  ConversionContext with 3-tier priority resolution (config override >\n  model override > provider default)\n- `apply_shim_to_ir(ir, shim)` — apply all IR-level shim transforms\n  i\n[…]\noth non-streaming and streaming handlers use identical one-liner calls.\nExternal consumers (e.g. argo-proxy) can now import from\n`llm_rosetta.shims` to get all shim-driven behavior with a single call.",
          "is_bot": false,
          "headline": "refactor: unify shim-driven IR transforms into shims/pipeline.py (#319)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-24T07:38:35Z",
          "body_truncated": true,
          "is_coding_agent": false
        },
        {
          "oid": "27865cd610f2c907f1e6ffd4c414b61dc37b3f8d",
          "body": "The v0.6.11 helpers/ reorganization moved sanitize_schema, the orphan-fix\nutilities, and the cache/schema/tool_content modules into the helpers/\nsubpackage, unintentionally breaking import paths that external callers\n(e.g. argo-proxy) relied on.\n\nRestore backward compatibility:\n- Re-export sanitize_\n[…]\nhema.py, base/tool_content.py,\n  base/cache.py that redirect to their new helpers/ locations\n\nCanonical path remains llm_rosetta.converters.base.helpers. Cache\nsingletons are shared across both paths.",
          "is_bot": false,
          "headline": "fix: restore legacy converters/base import paths (v0.6.12)",
          "author_name": "Peng Ding",
          "author_login": "Oaklight",
          "committed_at": "2026-06-23T13:50:55Z",
          "body_truncated": true,
          "is_coding_agent": false
        }
      ],
      "releases_count": 36,
      "commits_last_year": 715,
      "latest_release_at": "2026-07-20T08:30:07Z",
      "latest_release_tag": "v0.7.2",
      "releases_from_tags": false,
      "days_since_last_push": 1,
      "active_weeks_last_year": 27,
      "days_since_latest_release": 3,
      "mean_days_between_releases": 3.5
    },
    "community": {
      "has_readme": true,
      "has_license": true,
      "has_description": true,
      "has_contributing": true,
      "health_percentage": 57,
      "has_issue_template": false,
      "has_code_of_conduct": false,
      "has_pull_request_template": false
    },
    "ecosystem": {
      "packages": [
        {
          "name": "llm-rosetta",
          "exists": true,
          "license": "MIT",
          "keywords": [
            "openai",
            "anthropic",
            "gemini",
            "google-genai",
            "api-adapter",
            "llm-gateway",
            "llmir",
            "streaming",
            "tool-calling",
            "multi-provider",
            "responses-api",
            "Development Status :: 3 - Alpha",
            "Intended Audience :: Developers",
            "Programming Language :: Python :: 3",
            "Programming Language :: Python :: 3.10",
            "Programming Language :: Python :: 3.11",
            "Programming Language :: Python :: 3.12",
            "Programming Language :: Python :: 3.13",
            "Topic :: Scientific/Engineering :: Artificial Intelligence",
            "Topic :: Software Development :: Libraries :: Python Modules"
          ],
          "ecosystem": "pypi",
          "matches_repo": true,
          "registry_url": "https://pypi.org/project/llm-rosetta/",
          "is_deprecated": false,
          "latest_version": "0.7.2",
          "repository_url": "https://github.com/Oaklight/llm-rosetta",
          "versions_count": 36,
          "total_downloads": null,
          "dependents_count": null,
          "deprecation_note": null,
          "maintainers_count": null,
          "monthly_downloads": 4817,
          "first_published_at": "2026-03-15T08:24:55.970511Z",
          "latest_published_at": "2026-07-20T08:29:54.948688Z",
          "latest_version_yanked": null,
          "days_since_latest_publish": 3
        }
      ]
    },
    "popularity": {
      "forks": 5,
      "stars": 58,
      "watchers": 1,
      "fork_history": {
        "days": [
          {
            "date": "2026-03-22",
            "count": 1
          },
          {
            "date": "2026-06-03",
            "count": 1
          },
          {
            "date": "2026-06-23",
            "count": 1
          },
          {
            "date": "2026-07-04",
            "count": 1
          },
          {
            "date": "2026-07-06",
            "count": 1
          }
        ],
        "complete": true,
        "collected": 5,
        "total_forks": 5
      },
      "star_history": null,
      "open_issues_and_prs": 31
    },
    "ai_readiness": {
      "has_nix": false,
      "example_dirs": [
        "examples"
      ],
      "has_llms_txt": false,
      "has_dockerfile": true,
      "has_mcp_signal": false,
      "bootstrap_files": [
        "Makefile"
      ],
      "api_schema_files": [],
      "has_devcontainer": false,
      "typecheck_configs": [
        "src/llm_rosetta/py.typed"
      ],
      "toolchain_manifests": [],
      "largest_source_bytes": 91915,
      "source_files_sampled": 297,
      "oversized_source_files": 4,
      "agent_instruction_files": [
        "AGENTS.md",
        "CLAUDE.md"
      ],
      "agent_instruction_max_bytes": 11571
    },
    "dependencies": {
      "manifests": [
        "pyproject.toml"
      ],
      "advisories": {
        "error": null,
        "scope": "repository_graph",
        "source": "osv",
        "findings": [
          {
            "name": "cryptography",
            "direct": false,
            "version": "48.0.0",
            "severity": "high",
            "ecosystem": "pypi",
            "cvss_score": 7.5,
            "advisory_ids": [
              "GHSA-537c-gmf6-5ccf"
            ],
            "fixed_version": "48.0.1",
            "advisory_count": 1,
            "oldest_advisory_days": 37
          },
          {
            "name": "pyasn1",
            "direct": false,
            "version": "0.6.3",
            "severity": "high",
            "ecosystem": "pypi",
            "cvss_score": 7.5,
            "advisory_ids": [
              "GHSA-8ppf-4f7h-5ppj",
              "GHSA-hm4w-wwcw-mr6r",
              "PYSEC-2026-3455",
              "PYSEC-2026-3456",
              "PYSEC-2026-3457"
            ],
            "fixed_version": "0.6.4",
            "advisory_count": 5,
            "oldest_advisory_days": 8
          }
        ],
        "collected": true,
        "malicious": [],
        "truncated": false,
        "by_severity": {
          "high": 2
        },
        "advisory_count": 6,
        "affected_count": 2,
        "assessed_count": 82,
        "malicious_count": 0,
        "assessed_package": null,
        "unassessed_count": 13,
        "direct_affected_count": 0
      },
      "ecosystems": [
        "pypi"
      ],
      "dependencies": [
        {
          "name": "typing_extensions",
          "manifest": "pyproject.toml",
          "ecosystem": "pypi",
          "version_constraint": ">=4.0.0"
        }
      ],
      "all_dependencies": {
        "error": null,
        "source": "github-sbom",
        "packages": [
          {
            "name": "typing-extensions",
            "direct": true,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "typing-extensions",
            "direct": true,
            "version": "4.15.0",
            "ecosystem": "pypi"
          },
          {
            "name": "annotated-doc",
            "direct": false,
            "version": "0.0.4",
            "ecosystem": "pypi"
          },
          {
            "name": "annotated-types",
            "direct": false,
            "version": "0.7.0",
            "ecosystem": "pypi"
          },
          {
            "name": "anthropic",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "anthropic",
            "direct": false,
            "version": "0.105.2",
            "ecosystem": "pypi"
          },
          {
            "name": "anyio",
            "direct": false,
            "version": "4.13.0",
            "ecosystem": "pypi"
          },
          {
            "name": "backports-tarfile",
            "direct": false,
            "version": "1.2.0",
            "ecosystem": "pypi"
          },
          {
            "name": "build",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "build",
            "direct": false,
            "version": "1.5.0",
            "ecosystem": "pypi"
          },
          {
            "name": "certifi",
            "direct": false,
            "version": "2026.5.20",
            "ecosystem": "pypi"
          },
          {
            "name": "cffi",
            "direct": false,
            "version": "2.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "cfgv",
            "direct": false,
            "version": "3.5.0",
            "ecosystem": "pypi"
          },
          {
            "name": "charset-normalizer",
            "direct": false,
            "version": "3.4.7",
            "ecosystem": "pypi"
          },
          {
            "name": "colorama",
            "direct": false,
            "version": "0.4.6",
            "ecosystem": "pypi"
          },
          {
            "name": "complexipy",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "complexipy",
            "direct": false,
            "version": "5.5.0",
            "ecosystem": "pypi"
          },
          {
            "name": "coverage",
            "direct": false,
            "version": "7.14.1",
            "ecosystem": "pypi"
          },
          {
            "name": "cryptography",
            "direct": false,
            "version": "48.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "distlib",
            "direct": false,
            "version": "0.4.1",
            "ecosystem": "pypi"
          },
          {
            "name": "distro",
            "direct": false,
            "version": "1.9.0",
            "ecosystem": "pypi"
          },
          {
            "name": "docstring-parser",
            "direct": false,
            "version": "0.18.0",
            "ecosystem": "pypi"
          },
          {
            "name": "docutils",
            "direct": false,
            "version": "0.23",
            "ecosystem": "pypi"
          },
          {
            "name": "exceptiongroup",
            "direct": false,
            "version": "1.3.1",
            "ecosystem": "pypi"
          },
          {
            "name": "filelock",
            "direct": false,
            "version": "3.29.1",
            "ecosystem": "pypi"
          },
          {
            "name": "google-auth",
            "direct": false,
            "version": "2.53.0",
            "ecosystem": "pypi"
          },
          {
            "name": "google-genai",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "google-genai",
            "direct": false,
            "version": "2.8.0",
            "ecosystem": "pypi"
          },
          {
            "name": "h11",
            "direct": false,
            "version": "0.16.0",
            "ecosystem": "pypi"
          },
          {
            "name": "httpcore",
            "direct": false,
            "version": "1.0.9",
            "ecosystem": "pypi"
          },
          {
            "name": "httpx",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "httpx",
            "direct": false,
            "version": "0.28.1",
            "ecosystem": "pypi"
          },
          {
            "name": "id",
            "direct": false,
            "version": "1.6.1",
            "ecosystem": "pypi"
          },
          {
            "name": "identify",
            "direct": false,
            "version": "2.6.19",
            "ecosystem": "pypi"
          },
          {
            "name": "idna",
            "direct": false,
            "version": "3.18",
            "ecosystem": "pypi"
          },
          {
            "name": "importlib-metadata",
            "direct": false,
            "version": "9.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "iniconfig",
            "direct": false,
            "version": "2.3.0",
            "ecosystem": "pypi"
          },
          {
            "name": "jaraco-classes",
            "direct": false,
            "version": "3.4.0",
            "ecosystem": "pypi"
          },
          {
            "name": "jaraco-context",
            "direct": false,
            "version": "6.1.2",
            "ecosystem": "pypi"
          },
          {
            "name": "jaraco-functools",
            "direct": false,
            "version": "4.5.0",
            "ecosystem": "pypi"
          },
          {
            "name": "jeepney",
            "direct": false,
            "version": "0.9.0",
            "ecosystem": "pypi"
          },
          {
            "name": "jiter",
            "direct": false,
            "version": "0.15.0",
            "ecosystem": "pypi"
          },
          {
            "name": "keyring",
            "direct": false,
            "version": "25.7.0",
            "ecosystem": "pypi"
          },
          {
            "name": "llm-rosetta",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "markdown-it-py",
            "direct": false,
            "version": "4.2.0",
            "ecosystem": "pypi"
          },
          {
            "name": "mdurl",
            "direct": false,
            "version": "0.1.2",
            "ecosystem": "pypi"
          },
          {
            "name": "more-itertools",
            "direct": false,
            "version": "11.1.0",
            "ecosystem": "pypi"
          },
          {
            "name": "nh3",
            "direct": false,
            "version": "0.3.5",
            "ecosystem": "pypi"
          },
          {
            "name": "nodeenv",
            "direct": false,
            "version": "1.10.0",
            "ecosystem": "pypi"
          },
          {
            "name": "openai",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "openai",
            "direct": false,
            "version": "2.41.0",
            "ecosystem": "pypi"
          },
          {
            "name": "packaging",
            "direct": false,
            "version": "26.2",
            "ecosystem": "pypi"
          },
          {
            "name": "platformdirs",
            "direct": false,
            "version": "4.10.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pluggy",
            "direct": false,
            "version": "1.6.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pre-commit",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "pre-commit",
            "direct": false,
            "version": "4.6.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pyasn1",
            "direct": false,
            "version": "0.6.3",
            "ecosystem": "pypi"
          },
          {
            "name": "pyasn1-modules",
            "direct": false,
            "version": "0.4.2",
            "ecosystem": "pypi"
          },
          {
            "name": "pycparser",
            "direct": false,
            "version": "3.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pydantic",
            "direct": false,
            "version": "2.13.4",
            "ecosystem": "pypi"
          },
          {
            "name": "pydantic-core",
            "direct": false,
            "version": "2.46.4",
            "ecosystem": "pypi"
          },
          {
            "name": "pygments",
            "direct": false,
            "version": "2.20.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pyinstrument",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "pyproject-hooks",
            "direct": false,
            "version": "1.2.0",
            "ecosystem": "pypi"
          },
          {
            "name": "pytest",
            "direct": false,
            "version": "9.0.3",
            "ecosystem": "pypi"
          },
          {
            "name": "pytest-cov",
            "direct": false,
            "version": "7.1.0",
            "ecosystem": "pypi"
          },
          {
            "name": "python-discovery",
            "direct": false,
            "version": "1.4.0",
            "ecosystem": "pypi"
          },
          {
            "name": "python-dotenv",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "python-dotenv",
            "direct": false,
            "version": "1.2.2",
            "ecosystem": "pypi"
          },
          {
            "name": "pywin32-ctypes",
            "direct": false,
            "version": "0.2.3",
            "ecosystem": "pypi"
          },
          {
            "name": "pyyaml",
            "direct": false,
            "version": "6.0.3",
            "ecosystem": "pypi"
          },
          {
            "name": "readme-renderer",
            "direct": false,
            "version": "44.0",
            "ecosystem": "pypi"
          },
          {
            "name": "requests",
            "direct": false,
            "version": "2.34.2",
            "ecosystem": "pypi"
          },
          {
            "name": "requests-toolbelt",
            "direct": false,
            "version": "1.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "rfc3986",
            "direct": false,
            "version": "2.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "rich",
            "direct": false,
            "version": "15.0.0",
            "ecosystem": "pypi"
          },
          {
            "name": "ruff",
            "direct": false,
            "version": "0.15.16",
            "ecosystem": "pypi"
          },
          {
            "name": "ruff",
            "direct": false,
            "version": "0.15.20",
            "ecosystem": "pypi"
          },
          {
            "name": "secretstorage",
            "direct": false,
            "version": "3.5.0",
            "ecosystem": "pypi"
          },
          {
            "name": "setuptools",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "shellingham",
            "direct": false,
            "version": "1.5.4",
            "ecosystem": "pypi"
          },
          {
            "name": "sniffio",
            "direct": false,
            "version": "1.3.1",
            "ecosystem": "pypi"
          },
          {
            "name": "tenacity",
            "direct": false,
            "version": "9.1.4",
            "ecosystem": "pypi"
          },
          {
            "name": "tomli",
            "direct": false,
            "version": "2.4.1",
            "ecosystem": "pypi"
          },
          {
            "name": "tqdm",
            "direct": false,
            "version": "4.67.3",
            "ecosystem": "pypi"
          },
          {
            "name": "twine",
            "direct": false,
            "version": null,
            "ecosystem": "pypi"
          },
          {
            "name": "twine",
            "direct": false,
            "version": "6.2.0",
            "ecosystem": "pypi"
          },
          {
            "name": "ty",
            "direct": false,
            "version": "0.0.34",
            "ecosystem": "pypi"
          },
          {
            "name": "ty",
            "direct": false,
            "version": "0.0.54",
            "ecosystem": "pypi"
          },
          {
            "name": "typer",
            "direct": false,
            "version": "0.26.7",
            "ecosystem": "pypi"
          },
          {
            "name": "typing-inspection",
            "direct": false,
            "version": "0.4.2",
            "ecosystem": "pypi"
          },
          {
            "name": "urllib3",
            "direct": false,
            "version": "2.7.0",
            "ecosystem": "pypi"
          },
          {
            "name": "virtualenv",
            "direct": false,
            "version": "21.4.2",
            "ecosystem": "pypi"
          },
          {
            "name": "websockets",
            "direct": false,
            "version": "16.0",
            "ecosystem": "pypi"
          },
          {
            "name": "zipp",
            "direct": false,
            "version": "4.1.0",
            "ecosystem": "pypi"
          }
        ],
        "collected": true,
        "truncated": false,
        "total_count": 95,
        "direct_count": 2,
        "indirect_count": 93
      }
    },
    "maintainership": {
      "issues": {
        "open_prs": 2,
        "merged_prs": 186,
        "open_issues": 29,
        "closed_ratio": 0.831,
        "closed_issues": 143,
        "closed_unmerged_prs": 24
      },
      "bus_factor": 1,
      "bot_contributors": 4,
      "top_contributors": [
        {
          "type": "User",
          "login": "Oaklight",
          "commits": 674,
          "avatar_url": "https://avatars.githubusercontent.com/u/13076552?v=4"
        },
        {
          "type": "User",
          "login": "caidao22",
          "commits": 2,
          "avatar_url": "https://avatars.githubusercontent.com/u/6456049?v=4"
        }
      ],
      "contributors_sampled": 2,
      "top_contributor_share": 0.997
    },
    "quality_signals": {
      "has_ci": true,
      "has_tests": true,
      "ci_workflows": [
        "ci.yml",
        "docker-safety.yml",
        "docker.yml",
        "release.yml",
        "sdk-compatibility.yml",
        "sdk-type-scan.yml"
      ],
      "has_docs_dir": false,
      "linter_configs": [],
      "has_editorconfig": false,
      "has_linter_config": true,
      "has_precommit_config": true
    },
    "security_signals": {
      "lockfiles": [],
      "scorecard": {
        "checks": [
          {
            "name": "Binary-Artifacts",
            "score": 10,
            "reason": "no binaries found in the repo",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#binary-artifacts"
          },
          {
            "name": "Branch-Protection",
            "score": 0,
            "reason": "branch protection not enabled on development/release branches",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#branch-protection"
          },
          {
            "name": "CI-Tests",
            "score": 10,
            "reason": "7 out of 7 merged PRs checked by a CI test -- score normalized to 10",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#ci-tests"
          },
          {
            "name": "CII-Best-Practices",
            "score": 0,
            "reason": "no effort to earn an OpenSSF best practices badge detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#cii-best-practices"
          },
          {
            "name": "Code-Review",
            "score": 1,
            "reason": "Found 2/20 approved changesets -- score normalized to 1",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#code-review"
          },
          {
            "name": "Contributors",
            "score": 3,
            "reason": "project has 1 contributing companies or organizations -- score normalized to 3",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#contributors"
          },
          {
            "name": "Dangerous-Workflow",
            "score": 10,
            "reason": "no dangerous workflow patterns detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dangerous-workflow"
          },
          {
            "name": "Dependency-Update-Tool",
            "score": 10,
            "reason": "update tool detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dependency-update-tool"
          },
          {
            "name": "Fuzzing",
            "score": 0,
            "reason": "project is not fuzzed",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#fuzzing"
          },
          {
            "name": "License",
            "score": 10,
            "reason": "license file detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#license"
          },
          {
            "name": "Maintained",
            "score": 10,
            "reason": "30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#maintained"
          },
          {
            "name": "Packaging",
            "score": 10,
            "reason": "packaging workflow detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#packaging"
          },
          {
            "name": "Pinned-Dependencies",
            "score": 0,
            "reason": "dependency not pinned by hash detected -- score normalized to 0",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#pinned-dependencies"
          },
          {
            "name": "SAST",
            "score": 0,
            "reason": "SAST tool is not run on all commits -- score normalized to 0",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#sast"
          },
          {
            "name": "Security-Policy",
            "score": 0,
            "reason": "security policy file not detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#security-policy"
          },
          {
            "name": "Signed-Releases",
            "score": 0,
            "reason": "Project has not signed or included provenance with any releases.",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#signed-releases"
          },
          {
            "name": "Token-Permissions",
            "score": 0,
            "reason": "detected GitHub workflow tokens with excessive permissions",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#token-permissions"
          },
          {
            "name": "Vulnerabilities",
            "score": 10,
            "reason": "0 existing vulnerabilities detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#vulnerabilities"
          }
        ],
        "commit": "4b268d1a7e9d014340bead0fa4b121a462a0708a",
        "ran_at": "2026-07-23T09:25:05Z",
        "aggregate_score": 4.9,
        "scorecard_version": "v5.5.0"
      },
      "has_codeql_workflow": false,
      "has_security_policy": false,
      "has_dependabot_config": true
    },
    "contribution_flow": {
      "collected": true,
      "ci_last_run_at": "2026-07-21T14:34:12Z",
      "oldest_open_prs": [
        {
          "number": 354,
          "created_at": "2026-07-06T21:27:56Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 380,
          "created_at": "2026-07-20T10:11:29Z",
          "last_comment_at": null,
          "last_comment_author": null
        }
      ],
      "last_merged_pr_at": "2026-07-21T14:33:29Z",
      "ci_last_conclusion": "SUCCESS",
      "oldest_open_issues": [
        {
          "number": 40,
          "created_at": "2026-03-15T14:55:42Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 41,
          "created_at": "2026-03-15T14:56:11Z",
          "last_comment_at": "2026-06-05T06:35:33Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 42,
          "created_at": "2026-03-15T14:56:18Z",
          "last_comment_at": "2026-06-05T06:35:34Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 74,
          "created_at": "2026-03-21T09:54:44Z",
          "last_comment_at": "2026-04-04T17:14:34Z",
          "last_comment_author": "Oaklight"
        },
        {
          "number": 75,
          "created_at": "2026-03-21T10:03:42Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 121,
          "created_at": "2026-04-09T15:39:27Z",
          "last_comment_at": "2026-06-05T06:35:14Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 123,
          "created_at": "2026-04-09T15:39:33Z",
          "last_comment_at": "2026-06-19T10:52:07Z",
          "last_comment_author": "elena-oaklight"
        },
        {
          "number": 124,
          "created_at": "2026-04-09T15:39:59Z",
          "last_comment_at": "2026-06-05T06:35:17Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 126,
          "created_at": "2026-04-09T15:40:05Z",
          "last_comment_at": "2026-06-19T10:52:08Z",
          "last_comment_author": "elena-oaklight"
        },
        {
          "number": 129,
          "created_at": "2026-04-09T15:40:43Z",
          "last_comment_at": "2026-06-05T06:35:15Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 130,
          "created_at": "2026-04-09T15:40:46Z",
          "last_comment_at": "2026-06-05T06:35:16Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 131,
          "created_at": "2026-04-09T15:40:48Z",
          "last_comment_at": "2026-06-05T06:35:36Z",
          "last_comment_author": "milo-oaklight"
        },
        {
          "number": 181,
          "created_at": "2026-05-09T16:10:51Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 185,
          "created_at": "2026-05-15T10:41:36Z",
          "last_comment_at": "2026-06-10T07:17:04Z",
          "last_comment_author": "Oaklight"
        },
        {
          "number": 201,
          "created_at": "2026-05-17T08:42:09Z",
          "last_comment_at": "2026-06-19T10:52:26Z",
          "last_comment_author": "elena-oaklight"
        },
        {
          "number": 217,
          "created_at": "2026-05-20T16:58:06Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 218,
          "created_at": "2026-05-22T12:57:38Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 220,
          "created_at": "2026-05-23T21:41:54Z",
          "last_comment_at": "2026-06-09T09:24:01Z",
          "last_comment_author": "Oaklight"
        },
        {
          "number": 270,
          "created_at": "2026-06-10T09:22:59Z",
          "last_comment_at": null,
          "last_comment_author": null
        },
        {
          "number": 271,
          "created_at": "2026-06-11T01:26:27Z",
          "last_comment_at": "2026-07-10T03:09:20Z",
          "last_comment_author": "jerry-yuan"
        }
      ]
    }
  },
  "config": {
    "disabled_metrics": [],
    "disabled_categories": [],
    "disabled_components": {}
  },
  "source": {
    "url": "https://github.com/Oaklight/llm-rosetta",
    "host": "github.com",
    "name": "llm-rosetta",
    "owner": "Oaklight"
  },
  "metrics": {
    "overall": {
      "key": "overall",
      "band": "moderate",
      "name": "Overall health",
      "note": null,
      "notes": [],
      "value": 68,
      "inputs": {
        "security": 59,
        "vitality": 86,
        "community": 53,
        "governance": 58,
        "engineering": 82
      },
      "components": []
    },
    "categories": [
      {
        "key": "vitality",
        "band": "excellent",
        "name": "Vitality",
        "value": 86,
        "weight": 0.22,
        "metrics": [
          {
            "key": "development_activity",
            "band": "good",
            "name": "Development activity",
            "note": null,
            "notes": [],
            "value": 83,
            "inputs": {
              "commits_last_year": 715,
              "human_commit_share": 0.93,
              "days_since_last_push": 1,
              "active_weeks_last_year": 27
            },
            "components": [
              {
                "key": "push_recency",
                "name": "Push recency",
                "detail": "last push 1 days ago",
                "points": 36,
                "status": "met",
                "details": [
                  {
                    "code": "push_recency",
                    "params": {
                      "days": 1
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "commit_cadence",
                "name": "Commit cadence",
                "detail": "27/52 weeks with commits",
                "points": 18.7,
                "status": "partial",
                "details": [
                  {
                    "code": "commit_cadence_weeks",
                    "params": {
                      "weeks": 27
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "commit_volume",
                "name": "Commit volume",
                "detail": "715 commits in the last year",
                "points": 18,
                "status": "met",
                "details": [
                  {
                    "code": "commits_last_year",
                    "params": {
                      "count": 715
                    }
                  }
                ],
                "max_points": 18
              },
              {
                "key": "openssf_scorecard_maintained",
                "name": "OpenSSF Scorecard: Maintained",
                "detail": "30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10",
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "release_discipline",
            "band": "excellent",
            "name": "Release discipline",
            "note": null,
            "notes": [],
            "value": 90,
            "inputs": {
              "releases_count": 36,
              "latest_release_tag": "v0.7.2",
              "releases_from_tags": false,
              "days_since_latest_release": 3,
              "mean_days_between_releases": 3.5
            },
            "components": [
              {
                "key": "ships_releases",
                "name": "Ships releases",
                "detail": "36 releases published",
                "points": 27,
                "status": "met",
                "details": [
                  {
                    "code": "releases_published",
                    "params": {
                      "count": 36
                    }
                  }
                ],
                "max_points": 27
              },
              {
                "key": "release_recency",
                "name": "Release recency",
                "detail": "latest release 3 days ago",
                "points": 36,
                "status": "met",
                "details": [
                  {
                    "code": "release_recency",
                    "params": {
                      "days": 3
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "release_cadence",
                "name": "Release cadence",
                "detail": "a release every ~3.5 days",
                "points": 27,
                "status": "met",
                "details": [
                  {
                    "code": "release_cadence",
                    "params": {
                      "gap": 3.5
                    }
                  }
                ],
                "max_points": 27
              },
              {
                "key": "openssf_scorecard_signed_releases",
                "name": "OpenSSF Scorecard: Signed-Releases",
                "detail": "Project has not signed or included provenance with any releases.",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "abandonment",
            "band": "excellent",
            "name": "Abandonment",
            "note": null,
            "notes": [],
            "value": 100,
            "inputs": {
              "cap": null,
              "state": "maintained",
              "guards": [],
              "signals": [],
              "red_flag": false,
              "multiplier_pct": 100,
              "declared_reason": null,
              "unverified_reason": null,
              "unanswered_open_prs": null,
              "unanswered_open_issues": null,
              "days_since_last_merged_pr": null,
              "days_since_last_human_commit": 1,
              "days_since_last_human_commit_is_floor": false
            },
            "components": [
              {
                "key": "project_is_still_maintained",
                "name": "Project is still maintained",
                "detail": "last human commit 1 days ago",
                "points": 100,
                "status": "met",
                "details": [
                  {
                    "code": "abandonment_maintained",
                    "params": {
                      "days": 1
                    }
                  }
                ],
                "max_points": 100
              }
            ]
          }
        ],
        "description": "Is the project alive — is code being written and are releases shipping?"
      },
      {
        "key": "community",
        "band": "moderate",
        "name": "Community & Adoption",
        "value": 53,
        "weight": 0.18,
        "metrics": [
          {
            "key": "popularity",
            "band": "at_risk",
            "name": "Popularity & adoption",
            "note": null,
            "notes": [],
            "value": 34,
            "inputs": {
              "forks": 5,
              "stars": 58,
              "watchers": 1,
              "growth_state": "unverified",
              "growth_factor_pct": 100,
              "growth_unverified_reason": "no_history"
            },
            "components": [
              {
                "key": "stars",
                "name": "Stars",
                "detail": "58 stars",
                "points": 28.5,
                "status": "partial",
                "details": [
                  {
                    "code": "stars",
                    "params": {
                      "count": 58
                    }
                  }
                ],
                "max_points": 60
              },
              {
                "key": "forks",
                "name": "Forks",
                "detail": "5 forks",
                "points": 5,
                "status": "partial",
                "details": [
                  {
                    "code": "forks",
                    "params": {
                      "count": 5
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "watchers",
                "name": "Watchers",
                "detail": "1 watchers",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "watchers",
                    "params": {
                      "count": 1
                    }
                  }
                ],
                "max_points": 15
              }
            ]
          },
          {
            "key": "community_health",
            "band": "good",
            "name": "Community health",
            "note": null,
            "notes": [],
            "value": 70,
            "inputs": {
              "has_readme": true,
              "has_license": true,
              "has_contributing": true,
              "has_issue_template": false,
              "has_code_of_conduct": false,
              "has_pull_request_template": false
            },
            "components": [
              {
                "key": "readme",
                "name": "README",
                "detail": null,
                "points": 22.5,
                "status": "met",
                "details": [],
                "max_points": 22.5
              },
              {
                "key": "license",
                "name": "License",
                "detail": "recognized license (MIT)",
                "points": 22.5,
                "status": "met",
                "details": [
                  {
                    "code": "license_standard",
                    "params": {}
                  },
                  {
                    "code": "license_spdx",
                    "params": {
                      "spdx": "MIT"
                    }
                  }
                ],
                "max_points": 22.5
              },
              {
                "key": "contributing_guide",
                "name": "CONTRIBUTING guide",
                "detail": null,
                "points": 18,
                "status": "met",
                "details": [],
                "max_points": 18
              },
              {
                "key": "code_of_conduct",
                "name": "Code of conduct",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 13.5
              },
              {
                "key": "issue_template",
                "name": "Issue template",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.2
              },
              {
                "key": "pr_template",
                "name": "PR template",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 6.3
              }
            ]
          },
          {
            "key": "ecosystem_adoption",
            "band": "moderate",
            "name": "Ecosystem adoption (downloads)",
            "note": "Excluded from scoring (no data or not applicable): Registry dependents. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "registry_dependents"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 61,
            "inputs": {
              "packages": [
                "llm-rosetta"
              ],
              "dependents": null,
              "ecosystems": "pypi",
              "total_downloads": null,
              "monthly_downloads": 4817
            },
            "components": [
              {
                "key": "monthly_downloads",
                "name": "Monthly downloads",
                "detail": "4,817 downloads/month across pypi",
                "points": 49.1,
                "status": "partial",
                "details": [
                  {
                    "code": "downloads_monthly",
                    "params": {
                      "count": 4817,
                      "ecosystems": "pypi"
                    }
                  }
                ],
                "max_points": 80
              },
              {
                "key": "registry_dependents",
                "name": "Registry dependents",
                "detail": "not reported by this ecosystem",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "not_reported_by_this_ecosystem",
                    "params": {}
                  }
                ],
                "max_points": 20
              }
            ]
          }
        ],
        "description": "Does the project have users, downloads, attention, and a welcoming setup for contributors?"
      },
      {
        "key": "governance",
        "band": "moderate",
        "name": "Sustainability & Governance",
        "value": 58,
        "weight": 0.24,
        "metrics": [
          {
            "key": "maintainer_resilience",
            "band": "critical",
            "name": "Maintainer resilience (bus factor)",
            "note": null,
            "notes": [],
            "value": 15,
            "inputs": {
              "bus_factor": 1,
              "contributors_sampled": 2,
              "top_contributor_share": 0.997
            },
            "components": [
              {
                "key": "bus_factor",
                "name": "Bus factor",
                "detail": "1 contributor(s) cover half of all commits",
                "points": 9,
                "status": "partial",
                "details": [
                  {
                    "code": "bus_factor",
                    "params": {
                      "count": 1
                    }
                  }
                ],
                "max_points": 54
              },
              {
                "key": "commit_distribution",
                "name": "Commit distribution",
                "detail": "top contributor authored 100% of commits",
                "points": 0.1,
                "status": "partial",
                "details": [
                  {
                    "code": "top_contributor_share",
                    "params": {
                      "share": 100
                    }
                  }
                ],
                "max_points": 22.5
              },
              {
                "key": "contributor_breadth",
                "name": "Contributor breadth",
                "detail": "2 contributors",
                "points": 2.7,
                "status": "partial",
                "details": [
                  {
                    "code": "contributors_sampled",
                    "params": {
                      "count": 2
                    }
                  }
                ],
                "max_points": 13.5
              },
              {
                "key": "openssf_scorecard_contributors",
                "name": "OpenSSF Scorecard: Contributors",
                "detail": "project has 1 contributing companies or organizations -- score normalized to 3",
                "points": 3,
                "status": "partial",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "responsiveness",
            "band": "good",
            "name": "Issue & PR responsiveness",
            "note": null,
            "notes": [],
            "value": 74,
            "inputs": {
              "merged_prs": 186,
              "open_issues": 29,
              "closed_issues": 143,
              "issue_closed_ratio": 0.831,
              "closed_unmerged_prs": 24
            },
            "components": [
              {
                "key": "issue_resolution",
                "name": "Issue resolution",
                "detail": "83% of issues closed",
                "points": 38.8,
                "status": "partial",
                "details": [
                  {
                    "code": "issues_closed_share",
                    "params": {
                      "share": 83
                    }
                  }
                ],
                "max_points": 46.75
              },
              {
                "key": "pr_acceptance",
                "name": "PR acceptance",
                "detail": "186/210 decided PRs merged",
                "points": 33.9,
                "status": "partial",
                "details": [
                  {
                    "code": "decided_prs_merged",
                    "params": {
                      "merged": 186,
                      "decided": 210
                    }
                  }
                ],
                "max_points": 38.25
              },
              {
                "key": "openssf_scorecard_code_review",
                "name": "OpenSSF Scorecard: Code-Review",
                "detail": "Found 2/20 approved changesets -- score normalized to 1",
                "points": 1.5,
                "status": "partial",
                "details": [],
                "max_points": 15
              }
            ]
          },
          {
            "key": "stewardship",
            "band": "moderate",
            "name": "Ownership & stewardship",
            "note": "Excluded from scoring (no data or not applicable): Verified domain. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "verified_domain"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 61,
            "inputs": {
              "followers": 78,
              "owner_type": "User",
              "is_verified": null,
              "owner_login": "Oaklight",
              "public_repos": 104,
              "account_age_days": 4043
            },
            "components": [
              {
                "key": "ownership_backing",
                "name": "Ownership backing",
                "detail": "personal (user) account",
                "points": 10,
                "status": "partial",
                "details": [
                  {
                    "code": "owner_personal",
                    "params": {}
                  }
                ],
                "max_points": 30
              },
              {
                "key": "verified_domain",
                "name": "Verified domain",
                "detail": "not applicable to user accounts",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "not_applicable_to_user_accounts",
                    "params": {}
                  }
                ],
                "max_points": 20
              },
              {
                "key": "owner_reach",
                "name": "Owner reach",
                "detail": "78 followers of Oaklight",
                "points": 13.6,
                "status": "partial",
                "details": [
                  {
                    "code": "owner_followers",
                    "params": {
                      "count": 78,
                      "login": "Oaklight"
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "track_record",
                "name": "Track record",
                "detail": "104 public repos, account ~11 yr old",
                "points": 25,
                "status": "met",
                "details": [
                  {
                    "code": "public_repos",
                    "params": {
                      "count": 104
                    }
                  },
                  {
                    "code": "account_age_years",
                    "params": {
                      "years": 11
                    }
                  }
                ],
                "max_points": 25
              }
            ]
          },
          {
            "key": "package_maintenance",
            "band": "excellent",
            "name": "Package maintenance",
            "note": null,
            "notes": [],
            "value": 100,
            "inputs": {
              "packages": [
                "llm-rosetta"
              ],
              "ecosystems": "pypi",
              "any_deprecated": false,
              "min_days_since_publish": 3
            },
            "components": [
              {
                "key": "published_resolvable",
                "name": "Published & resolvable",
                "detail": "1 package(s) on pypi",
                "points": 25,
                "status": "met",
                "details": [
                  {
                    "code": "packages_published",
                    "params": {
                      "count": 1,
                      "ecosystems": "pypi"
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "publish_recency",
                "name": "Publish recency",
                "detail": "latest publish 3 days ago",
                "points": 35,
                "status": "met",
                "details": [
                  {
                    "code": "publish_recency",
                    "params": {
                      "days": 3
                    }
                  }
                ],
                "max_points": 35
              },
              {
                "key": "version_history",
                "name": "Version history",
                "detail": "36 published versions",
                "points": 20,
                "status": "met",
                "details": [
                  {
                    "code": "published_versions",
                    "params": {
                      "count": 36
                    }
                  }
                ],
                "max_points": 20
              },
              {
                "key": "not_deprecated",
                "name": "Not deprecated",
                "detail": "active, not deprecated or yanked",
                "points": 20,
                "status": "met",
                "details": [
                  {
                    "code": "package_not_deprecated",
                    "params": {}
                  }
                ],
                "max_points": 20
              }
            ]
          }
        ],
        "description": "Will the project survive its people — bus factor, responsiveness, who backs it, and package upkeep?"
      },
      {
        "key": "engineering",
        "band": "good",
        "name": "Engineering Quality",
        "value": 82,
        "weight": 0.2,
        "metrics": [
          {
            "key": "engineering_practices",
            "band": "excellent",
            "name": "Engineering practices",
            "note": null,
            "notes": [],
            "value": 94,
            "inputs": {
              "has_ci": true,
              "has_tests": true,
              "has_editorconfig": false,
              "has_linter_config": true,
              "has_precommit_config": true
            },
            "components": [
              {
                "key": "ci_workflows",
                "name": "CI workflows",
                "detail": "6 workflow(s)",
                "points": 24,
                "status": "met",
                "details": [
                  {
                    "code": "ci_workflows",
                    "params": {
                      "count": 6
                    }
                  }
                ],
                "max_points": 24
              },
              {
                "key": "tests_present",
                "name": "Tests present",
                "detail": null,
                "points": 24,
                "status": "met",
                "details": [],
                "max_points": 24
              },
              {
                "key": "linter_config",
                "name": "Linter config",
                "detail": null,
                "points": 16,
                "status": "met",
                "details": [],
                "max_points": 16
              },
              {
                "key": "pre_commit_hooks",
                "name": "Pre-commit hooks",
                "detail": null,
                "points": 9.6,
                "status": "met",
                "details": [],
                "max_points": 9.6
              },
              {
                "key": "editorconfig",
                "name": ".editorconfig",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 6.4
              },
              {
                "key": "openssf_scorecard_ci_tests",
                "name": "OpenSSF Scorecard: CI-Tests",
                "detail": "7 out of 7 merged PRs checked by a CI test -- score normalized to 10",
                "points": 20,
                "status": "met",
                "details": [],
                "max_points": 20
              }
            ]
          },
          {
            "key": "documentation",
            "band": "moderate",
            "name": "Documentation",
            "note": null,
            "notes": [],
            "value": 65,
            "inputs": {
              "topics": [
                "anthropic",
                "gemini",
                "openai",
                "python",
                "streaming",
                "anthropic-api",
                "api-adapter",
                "llm-gateway",
                "openai-api",
                "llmir",
                "google-genai",
                "multi-provider",
                "responses-api",
                "tool-calling"
              ],
              "has_wiki": false,
              "homepage": "https://llm-rosetta.readthedocs.io",
              "has_readme": true,
              "has_docs_dir": false,
              "has_description": true
            },
            "components": [
              {
                "key": "readme",
                "name": "README",
                "detail": null,
                "points": 30,
                "status": "met",
                "details": [],
                "max_points": 30
              },
              {
                "key": "documentation_directory",
                "name": "Documentation directory",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 25
              },
              {
                "key": "documentation_homepage_site",
                "name": "Documentation / homepage site",
                "detail": "https://llm-rosetta.readthedocs.io",
                "points": 15,
                "status": "met",
                "details": [],
                "max_points": 15
              },
              {
                "key": "repository_description",
                "name": "Repository description",
                "detail": null,
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              },
              {
                "key": "topics",
                "name": "Topics",
                "detail": "14 topics",
                "points": 10,
                "status": "met",
                "details": [
                  {
                    "code": "topics_count",
                    "params": {
                      "count": 14
                    }
                  }
                ],
                "max_points": 10
              },
              {
                "key": "wiki",
                "name": "Wiki",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              }
            ]
          }
        ],
        "description": "Are baseline engineering and documentation practices in place?"
      },
      {
        "key": "security",
        "band": "moderate",
        "name": "Security",
        "value": 59,
        "weight": 0.16,
        "metrics": [
          {
            "key": "security_posture",
            "band": "at_risk",
            "name": "Security posture",
            "note": null,
            "notes": [],
            "value": 49,
            "inputs": {
              "source": "openssf_scorecard",
              "checks_evaluated": 18,
              "scorecard_version": "v5.5.0",
              "checks_inconclusive": 0,
              "scorecard_aggregate": 4.9
            },
            "components": [
              {
                "key": "binary_artifacts",
                "name": "Binary-Artifacts",
                "detail": "no binaries found in the repo",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "branch_protection",
                "name": "Branch-Protection",
                "detail": "branch protection not enabled on development/release branches",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "ci_tests",
                "name": "CI-Tests",
                "detail": "7 out of 7 merged PRs checked by a CI test -- score normalized to 10",
                "points": 2.5,
                "status": "met",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "cii_best_practices",
                "name": "CII-Best-Practices",
                "detail": "no effort to earn an OpenSSF best practices badge detected",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "code_review",
                "name": "Code-Review",
                "detail": "Found 2/20 approved changesets -- score normalized to 1",
                "points": 0.8,
                "status": "partial",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "contributors",
                "name": "Contributors",
                "detail": "project has 1 contributing companies or organizations -- score normalized to 3",
                "points": 0.8,
                "status": "partial",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "dangerous_workflow",
                "name": "Dangerous-Workflow",
                "detail": "no dangerous workflow patterns detected",
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              },
              {
                "key": "dependency_update_tool",
                "name": "Dependency-Update-Tool",
                "detail": "update tool detected",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "fuzzing",
                "name": "Fuzzing",
                "detail": "project is not fuzzed",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "license",
                "name": "License",
                "detail": "license file detected",
                "points": 2.5,
                "status": "met",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "maintained",
                "name": "Maintained",
                "detail": "30 commit(s) and 27 issue activity found in the last 90 days -- score normalized to 10",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "packaging",
                "name": "Packaging",
                "detail": "packaging workflow detected",
                "points": 5,
                "status": "met",
                "details": [],
                "max_points": 5
              },
              {
                "key": "pinned_dependencies",
                "name": "Pinned-Dependencies",
                "detail": "dependency not pinned by hash detected -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "sast",
                "name": "SAST",
                "detail": "SAST tool is not run on all commits -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "security_policy",
                "name": "Security-Policy",
                "detail": "security policy file not detected",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "signed_releases",
                "name": "Signed-Releases",
                "detail": "Project has not signed or included provenance with any releases.",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "token_permissions",
                "name": "Token-Permissions",
                "detail": "detected GitHub workflow tokens with excessive permissions",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "vulnerabilities",
                "name": "Vulnerabilities",
                "detail": "0 existing vulnerabilities detected",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              }
            ]
          },
          {
            "key": "dependency_advisories",
            "band": "excellent",
            "name": "Dependency advisories",
            "note": "Excluded from scoring (no data or not applicable): Indirect dependencies free of known advisories, No advisories left outstanding. Remaining weights renormalized. Matched 82 resolved dependencies against OSV; 13 could not be assessed (no resolved version, an unsupported ecosystem, or beyond the reported package list). This repository publishes no package the index resolves, so the repository dependency graph was assessed instead. That graph mixes development and test pins with shipped dependencies, so only the declared runtime dependencies are scored; transitive findings are reported as context and excluded from the score. Reachability is not analyzed.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "indirect_dependencies_free_of_known_advisories",
                    "no_advisories_left_outstanding"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              },
              {
                "code": "advisories_scope_repository",
                "params": {
                  "assessed": 82
                }
              },
              {
                "code": "advisories_unassessed",
                "params": {
                  "count": 13
                }
              },
              {
                "code": "advisories_repo_graph_caveat",
                "params": {}
              },
              {
                "code": "advisories_reachability",
                "params": {}
              }
            ],
            "value": 100,
            "inputs": {
              "source": "osv",
              "advisories": 6,
              "affected_packages": 2,
              "assessed_packages": 82,
              "unassessed_packages": 13,
              "affected_by_severity": "high 2",
              "direct_affected_packages": 0
            },
            "components": [
              {
                "key": "direct_dependencies_free_of_known_advisories",
                "name": "Direct dependencies free of known advisories",
                "detail": "no direct dependency carries a known advisory",
                "points": 35,
                "status": "met",
                "details": [
                  {
                    "code": "no_direct_advisories",
                    "params": {}
                  }
                ],
                "max_points": 35
              },
              {
                "key": "indirect_dependencies_free_of_known_advisories",
                "name": "Indirect dependencies free of known advisories",
                "detail": "transitive set not separable from development and test dependencies in this scope",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "advisories_scope_not_separable",
                    "params": {}
                  }
                ],
                "max_points": 25
              },
              {
                "key": "no_advisories_left_outstanding",
                "name": "No advisories left outstanding",
                "detail": "no advisory carries a publication date",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "advisories_no_publication_date",
                    "params": {}
                  }
                ],
                "max_points": 40
              }
            ]
          },
          {
            "key": "malicious_dependencies",
            "band": "excellent",
            "name": "Malicious dependencies",
            "note": null,
            "notes": [],
            "value": 100,
            "inputs": {
              "source": "osv",
              "meaning": "reported as a malicious package by the OpenSSF corpus; the remedy is removal or moving off the compromised name, never an upgrade of the same artifact. Versions the registry has since pulled are listed but not scored",
              "packages": [],
              "red_flag": false,
              "assessed_packages": 82,
              "malicious_packages": 0,
              "direct_malicious_packages": 0,
              "withdrawn_malicious_packages": 0,
              "installable_malicious_packages": 0
            },
            "components": [
              {
                "key": "no_dependency_reported_as_a_malicious_package",
                "name": "No dependency reported as a malicious package",
                "detail": "no dependency is reported as a malicious package",
                "points": 100,
                "status": "met",
                "details": [
                  {
                    "code": "no_malicious_dependencies",
                    "params": {}
                  }
                ],
                "max_points": 100
              }
            ]
          }
        ],
        "description": "Are visible security and supply-chain practices strong, with no malicious dependency and no unresolved high-risk jurisdiction exposure?"
      },
      {
        "key": "ai_readiness",
        "band": "good",
        "name": "AI Readiness",
        "value": 74,
        "weight": 0,
        "metrics": [
          {
            "key": "ai_agent_context",
            "band": "excellent",
            "name": "Agent context & guidance",
            "note": null,
            "notes": [],
            "value": 85,
            "inputs": {
              "has_llms_txt": false,
              "legible_history_share": 0.978,
              "agent_instruction_files": [
                "AGENTS.md",
                "CLAUDE.md"
              ],
              "agent_instruction_max_bytes": 11571
            },
            "components": [
              {
                "key": "agent_instructions",
                "name": "Agent instructions",
                "detail": "AGENTS.md, CLAUDE.md",
                "points": 45,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "AGENTS.md, CLAUDE.md"
                    }
                  }
                ],
                "max_points": 45
              },
              {
                "key": "machine_readable_docs_llms_txt",
                "name": "Machine-readable docs (llms.txt)",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 15
              },
              {
                "key": "legible_commit_history",
                "name": "Legible commit history",
                "detail": "91 of 93 human commits state their intent (structured subject or explanatory body)",
                "points": 40,
                "status": "met",
                "details": [
                  {
                    "code": "legible_history",
                    "params": {
                      "legible": 91,
                      "sampled": 93
                    }
                  }
                ],
                "max_points": 40
              }
            ]
          },
          {
            "key": "ai_verify_loop",
            "band": "good",
            "name": "Verify loop (build / test / typecheck)",
            "note": null,
            "notes": [],
            "value": 77,
            "inputs": {
              "has_nix": false,
              "has_tests": true,
              "lockfiles": [],
              "has_dockerfile": true,
              "typed_language": false,
              "bootstrap_files": [
                "Makefile"
              ],
              "has_devcontainer": false,
              "has_linter_config": true,
              "typecheck_configs": [
                "src/llm_rosetta/py.typed"
              ],
              "agent_commit_share": 0,
              "toolchain_manifests": [],
              "dependency_bot_commit_share": 0
            },
            "components": [
              {
                "key": "one_command_bootstrap",
                "name": "One-command bootstrap",
                "detail": "Makefile",
                "points": 18,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "Makefile"
                    }
                  }
                ],
                "max_points": 18
              },
              {
                "key": "automated_tests",
                "name": "Automated tests",
                "detail": null,
                "points": 22,
                "status": "met",
                "details": [],
                "max_points": 22
              },
              {
                "key": "lint_format_config",
                "name": "Lint / format config",
                "detail": null,
                "points": 11,
                "status": "met",
                "details": [],
                "max_points": 11
              },
              {
                "key": "static_type_checking",
                "name": "Static type checking",
                "detail": "src/llm_rosetta/py.typed",
                "points": 11,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "src/llm_rosetta/py.typed"
                    }
                  }
                ],
                "max_points": 11
              },
              {
                "key": "reproducible_environment",
                "name": "Reproducible environment",
                "detail": "Dockerfile",
                "points": 10,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "Dockerfile"
                    }
                  }
                ],
                "max_points": 10
              },
              {
                "key": "demonstrated_agent_practice",
                "name": "Demonstrated agent practice",
                "detail": "no agent-authored commits among the last 100",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "no_agent_authored_commits",
                    "params": {
                      "sampled": 100
                    }
                  }
                ],
                "max_points": 10
              },
              {
                "key": "automated_maintenance",
                "name": "Automated maintenance",
                "detail": "dependency automation configured, none observed in the sampled commits",
                "points": 5,
                "status": "partial",
                "details": [
                  {
                    "code": "dependency_bot_config_only",
                    "params": {}
                  }
                ],
                "max_points": 8
              },
              {
                "key": "openssf_scorecard_pinned_dependencies",
                "name": "OpenSSF Scorecard: Pinned-Dependencies",
                "detail": "dependency not pinned by hash detected -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "ai_code_legibility",
            "band": "good",
            "name": "Code legibility for models",
            "note": null,
            "notes": [],
            "value": 81,
            "inputs": {
              "primary_language": "Python",
              "largest_source_bytes": 91915,
              "source_files_sampled": 297,
              "oversized_source_files": 4
            },
            "components": [
              {
                "key": "type_checkable_code",
                "name": "Type-checkable code",
                "detail": "Python with type-check config (src/llm_rosetta/py.typed)",
                "points": 27,
                "status": "partial",
                "details": [
                  {
                    "code": "typecheck_config_language",
                    "params": {
                      "files": "src/llm_rosetta/py.typed",
                      "language": "Python"
                    }
                  }
                ],
                "max_points": 45
              },
              {
                "key": "manageable_file_sizes",
                "name": "Manageable file sizes",
                "detail": "4/297 source files over 60KB",
                "points": 54.3,
                "status": "partial",
                "details": [
                  {
                    "code": "oversized_source_files",
                    "params": {
                      "kb": 60,
                      "sampled": 297,
                      "oversized": 4
                    }
                  }
                ],
                "max_points": 55
              }
            ]
          },
          {
            "key": "ai_interfaces",
            "band": "at_risk",
            "name": "Machine-readable interfaces",
            "note": null,
            "notes": [],
            "value": 40,
            "inputs": {
              "example_dirs": [
                "examples"
              ],
              "has_mcp_signal": false,
              "api_schema_files": []
            },
            "components": [
              {
                "key": "api_schema_openapi_graphql_proto",
                "name": "API schema (OpenAPI/GraphQL/proto)",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 40
              },
              {
                "key": "mcp_server",
                "name": "MCP server",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 20
              },
              {
                "key": "runnable_examples",
                "name": "Runnable examples",
                "detail": "examples",
                "points": 40,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "examples"
                    }
                  }
                ],
                "max_points": 40
              }
            ]
          }
        ],
        "description": "How well is the repo equipped to be developed and maintained with AI coding agents? An independent, experimental badge — weight 0.0, so it is surfaced on its own and does not affect the overall health score."
      }
    ],
    "metrics_version": "1.13.0"
  },
  "warnings": [
    "Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token",
    "deps.dev does not index pypi:llm-rosetta@0.7.2; advisories assessed against the repository dependency graph instead"
  ],
  "report_type": "repository",
  "generated_at": "2026-07-23T09:25:19.235510Z",
  "schema_version": "0.27.0",
  "badge_url": "https://raw.githubusercontent.com/inspect-software/badges/main/v1/o/Oaklight/llm-rosetta.svg",
  "full_name": "Oaklight/llm-rosetta",
  "license_state": "standard",
  "license_spdx": "MIT"
}

评分是信号,而非担保。 评分反映的是 GitHub 上公开可见的实践——不是代码审计,也不是安全保证。

缺失数据将被剔除并重新归一化权重,绝不按零分计。方法论已版本化并公开:指标 v1.13.0、模式 v0.27.0—— 完整方法论 · 指标知识库.

单项结果在整体记录中的位置: 汇总统计PyPI.