🛠️ The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, learning loop, and witness-signed releases. Works with Claude Code, Codex, pi.dev, Hermes, OpenClaw, and RVM (hardware-isolated sandbox).
TypeScript · JavaScriptMIT★ 510 estrellas⑂ 58 forksdesde jun 2026Ver en GitHub ↗
ruvnet/metaharness tiene un índice de salud de 63 sobre 100, lo que lo sitúa en la banda Moderado. Su puntuación más alta es Engineering Quality (80/100) y la más baja, Security (44/100). Se actualizó por última vez hace 6 días. Una sola persona concentra la mayor parte del trabajo reciente.
63
global / 100
Moderado
Índice de salud del software
Las métricas se agrupan en categorías ponderadas sobre una escala de 1 a 100. El resultado global parte de su media; cuando la evidencia pública activa la Política de Jurisdicciones de Alto Riesgo, la calificación se ajusta y recibe el límite 49 (En riesgo). Preparación para IA queda fuera.
63
Excelente85-100Ejemplar; cumple prácticamente todos los criterios evaluados
Bueno70-84Saludable; carencias menores
Moderado50-69Aceptable con carencias notables; se recomienda revisión
En riesgo30-49Debilidades significativas; su adopción exige cautela
Crítico1-29Problemas graves (proyecto abandonado, un solo mantenedor, sin higiene)
Perfil de puntuación
Cada eje es una categoría. La forma importa más que la media: un proyecto sano llena toda la figura, mientras que un perfil de picos y cráteres indica que la fortaleza en una dimensión enmascara el riesgo en otra.
10.588 seguidores197 repositorios públicosdesde nov 2012Not a Bot
Este repositorio pertenece a una cuenta personal. Un proyecto con un único propietario conlleva más riesgo de continuidad que uno respaldado por una organización.
¿Hasta qué punto está el repositorio preparado para desarrollarse y mantenerse con agentes de codificación de IA? Es una insignia independiente y experimental — peso 0,0, de modo que se presenta por separado y no afecta a la puntuación de salud global.
Arranque con un solo comando — Cargo.toml, crates/kernel-napi/Cargo.toml, crates/kernel-wasm/Cargo.toml (convención del toolchain, sin ejecutor de tareas)
Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token
Could not fetch crates package 'ruflo-kernel' from its registry
Could not fetch crates package 'ruflo-kernel-napi' from its registry
Could not fetch crates package 'ruflo-kernel-wasm' from its registry
Could not fetch crates package 'poker-darwin' from its registry
Could not fetch crates package 'template-catalog' from its registry
npm package '@metaharness/codex' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring
npm package '@metaharness/coding' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring
npm package '@metaharness/devops' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring
GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository
Más detalle
Historial de estrellas y forks 0 ★ / 58 ⇿
0Estrellas
58Forks
1Versiones
Cuándo se añadió cada estrella y fork, recopilado de GitHub y agrupado por día. El crecimiento acumulado se sitúa justo encima de las adiciones diarias que lo componen, de modo que ambos se leen en conjunto: la acumulación orgánica sostenida no se parece en nada a un pico abrupto y efímero. Cuando esa diferencia es medible, se informa como autenticidad del crecimiento.
Mayor 0Menor 0Parche 1
OpenSSF Scorecard 4.4 / 10
4.4agregado
Evaluación de seguridad independiente y agnóstica en cuanto a herramientas, procedente del proyecto de código abierto OpenSSF Scorecard. Cada comprobación premia una práctica de seguridad, no la herramienta de un proveedor concreto. Las comprobaciones que Scorecard no pudo determinar se marcan como n/d y se excluyen de la puntuación de seguridad (nunca se cuentan como cero).Scorecard v5.5.0 · 2026-07-23 23:02 UTC
No fue posible recopilar el conjunto de dependencias resuelto para este informe: GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository
Informe JSON sin procesar legible por máquina
{
"data": {
"repo": {
"topics": [
"agent-harness",
"agent-scaffolding",
"agentic-ai",
"agentic-framework",
"autonomous-agents",
"claude-code",
"codex",
"hermes-agent",
"mcp",
"mcp-server",
"multi-agent",
"openai-codex",
"pi-dev",
"scaffold",
"scaffolding",
"agent-harness-generator",
"create-agent-harness",
"openclaw",
"rvm",
"sandboxing"
],
"is_fork": false,
"size_kb": 24266,
"has_wiki": false,
"homepage": "https://Cognitum.One",
"languages": {
"CSS": 1991,
"HCL": 8182,
"HTML": 50986,
"Rust": 237059,
"Shell": 711853,
"Python": 31117,
"Dockerfile": 549,
"JavaScript": 2449183,
"TypeScript": 3586860,
"Go Template": 186319
},
"pushed_at": "2026-07-17T22:47:22Z",
"created_at": "2026-06-13T18:29:11Z",
"owner_type": "User",
"updated_at": "2026-07-23T16:35:51Z",
"description": "🛠️ The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, learning loop, and witness-signed releases. Works with Claude Code, Codex, pi.dev, Hermes, OpenClaw, and RVM (hardware-isolated sandbox).",
"is_archived": false,
"is_disabled": false,
"license_spdx": "MIT",
"default_branch": "main",
"license_spdx_raw": "MIT",
"primary_language": "TypeScript",
"significant_languages": [
"TypeScript",
"JavaScript"
]
},
"owner": {
"blog": "Cognitum.One",
"name": "rUv",
"type": "User",
"login": "ruvnet",
"company": "Not a Bot",
"location": "0x",
"followers": 10588,
"avatar_url": "https://avatars.githubusercontent.com/u/2934394?v=4",
"created_at": "2012-11-30T22:08:25Z",
"is_verified": null,
"public_repos": 197,
"account_age_days": 4983
},
"license": {
"state": "standard",
"spdx_id": "MIT",
"raw_spdx": "MIT",
"file_present": true,
"scorecard_found": true,
"profile_has_license": true
},
"activity": {
"releases": [
{
"tag": "v0.1.3",
"kind": "patch",
"published_at": "2026-06-14T16:48:00Z"
}
],
"recent_commits": [
{
"oid": "072b95c0a74610de008dca5473343a81619cef20",
"body": "Meta-Proxy v0.4.1 (cognitum-one/meta-proxy#28) fixes a real bug: `meta-proxy\nlogin`/`logout` only ever wrote ~/.ruflo/proxy-config.toml on disk, which an\nalready-running daemon (config loaded once at boot) never observed until\nrestarted — a logout could even be silently undone the next time the\ndaem\n[…]\nhis package, ran `proxy install --yes` with a\nfresh HOME (no version override) against the real signed release channel —\ninstalls, checksum + Ed25519 signature verify, and the binary contains the\nfix.",
"is_bot": false,
"headline": "fix(proxy): bump pinned Meta-Proxy version to 0.4.1 (#145)",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-17T22:47:21Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "eb891907f55e31111ffa355e92ad5fc267b9da48",
"body": "feat(cli): signed Meta-Proxy routing and worktree policy",
"is_bot": false,
"headline": "Merge pull request #144 from ruvnet/feat/optional-meta-proxy-sidecar",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-17T19:01:09Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "0623003398824f4a0adac3405973619b61a34c54",
"body": null,
"is_bot": false,
"headline": "feat(proxy): route launched clients with worktree policies",
"author_name": "ruvnet",
"author_login": "ruvnet",
"committed_at": "2026-07-17T17:59:50Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "852028bb746cac6cc9bdafd02b17836b3f25b0f9",
"body": null,
"is_bot": false,
"headline": "feat(cli): add optional signed meta-proxy sidecar",
"author_name": "ruvnet",
"author_login": "ruvnet",
"committed_at": "2026-07-17T17:12:47Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "c86bc12a10b53b1f3e4afee03f69609802236eb3",
"body": null,
"is_bot": false,
"headline": "docs: add ADR-236 codex brain MCP integration (#140)",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-09T00:10:55Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "dbc25e0a11e72a2e0cc8ede1e0da69bc46dd1ee9",
"body": "Update ADR-235 for the published flywheel 0.1.7 replay fix and gate re-execution support.",
"is_bot": false,
"headline": "docs: update flywheel replay ADR status",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-08T20:01:03Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "139d79b5efd91a224d0fd8e1e10a2af371a60d63",
"body": "…demo (0.1.1) (#136)\n\nShips the audit-tooling layer for the interpretability stack (user-directed polish):\n\n- `workspace-probe` bin — a JSON-in/JSON-out CLI over @metaharness/workspace-lens + this package, for\n audit workflows that don't want to write TS (composes with jq/CI):\n diag <lens.json> \n[…]\n runtime-error paths) via a stdout\nspy; full workspace-probe suite 15/15; tsc clean; healthcheck version green (0.1.1, independently-versioned).\nREADME gains CLI + demo sections. bin added to `files`.",
"is_bot": false,
"headline": "feat(workspace-probe): CLI (readout/probe/grade-mutation) + runnable …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T14:11:48Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "96f1cfa71a30e73cae171f13de052252bcec5ac5",
"body": "…r HTTP (0.1.1) (#132)\n\n* feat(workspace-lens): WorkspaceLens.fromUrl + fromRegistry — load a fitted lens over HTTP (0.1.1)\n\nAdds the roadmap network-loading path to @metaharness/workspace-lens (ADR-238): fit a lens offline,\npublish the artifact, load it at startup.\n\n- WorkspaceLens.fromUrl(url, {fe\n[…]\nntly-versioned). Dependency-free (uses global fetch, Node 20+).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\n\n* fix(workspace-lens): linear trailing-slash strip in fromRegistry (ReDoS — CodeQL js/redos)",
"is_bot": false,
"headline": "feat(workspace-lens): fromUrl + fromRegistry — load a fitted lens ove…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T04:58:05Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6e7d741620aeeb00c581e3e5a4924a436610a00e",
"body": "…ode bridge for workspace-lens (ADR-239) (#131)\n\nA thin, pure companion to @metaharness/workspace-lens (ADR-238) that turns Jacobian-Lens interpretability\nreceipts into evaluation signal:\n\n- workspaceProbeScore(receipts, {driftThreshold}) → { n, score=cleanFraction, meanDrift, flagRate,\n criticalRa\n[…]\nsc clean; healthcheck 8/8. Wired into build-ordered phase-2 (after workspace-lens)\n+ the version-coherence allowlist; ADR-239. (The lockfile also gains the workspace-lens entry that #129\ndid not add.)",
"is_bot": false,
"headline": "feat(workspace-probe): @metaharness/workspace-probe — eval + Darwin-M…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T04:49:02Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e1c78caed29358effefedd47eae84181290fa04f",
"body": "…erpretability primitive (ADR-238) (#129)\n\nA runtime measurement primitive for the \"verbalizable workspace\" of open-weight LLMs (Anthropic,\n2026-07-06, \"Verbalizable Representations Form a Global Workspace\"). Reads lens_l(h)=unembed(J_l·h) into\nworkspace tokens, a layer trajectory, drift/entropy, ve\n[…]\ntaharness/* package on the next tag (non-private workspace package). README documents the triage\ndeployment tiers, circuit-breaker pattern, the acceptance test, and the fitting-is-external constraint.",
"is_bot": false,
"headline": "feat(workspace-lens): @metaharness/workspace-lens — Jacobian-Lens int…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T04:01:58Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "f11fa67454b5f5680af8b8535691f2eade3d6df4",
"body": "… merges for 6h (#128)\n\nThe CI matrix jobs had no `timeout-minutes`, so they inherit GitHub's 360-minute (6h) default. When a\nWindows `cargo test` ran slow/stuck, it left every open PR un-mergeable with no fail-fast signal — you\ncouldn't tell a hung job from a slow one, and nothing would end it shor\n[…]\nn:\n- rust: 90 (observed worst ~45min on Windows)\n- wasm: 45\n- node / pack-install: 30\n- bench: 20\n\nConfig-only; no gate is weakened (a timeout only bounds runtime). YAML validated (all 6 jobs parse).",
"is_bot": false,
"headline": "ci: add timeout-minutes to every matrix job so a hung job can't block…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T02:13:11Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "cfd373a63cd340ab460c12d7a56c8e85fa4d81ca",
"body": "… accept malformed witnesses (#4) (#127)\n\nBeta feedback #4 (mutation finding): verifyWitness's shape gate uses EXACT checks — public_key.length\n=== 64, signature.length === 128, schema === 1. The existing tests only cover UNDER-length keys/sigs\n('short') and schema 999, so a boundary mutant that rel\n[…]\nder `!== 64` → `< 64`, and the schema test FAILS under\n`!== 1` → `> 1`; both pass on the real code (then reverted).\n\nTest-only; no behaviour change. Full create-agent-harness suite 384/384; tsc clean.",
"is_bot": false,
"headline": "test(witness): pin the shape-gate boundaries so a relaxed bound can't…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T02:03:27Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "83ae846084bf0052eb574163aa683b062f1fade9",
"body": "…ng can't slip in (#4) (#126)\n\nBeta feedback #4 (mutation finding): render()/extractVarReferences use a deliberately strict var-name\ncharset — /\\{\\{\\s*([a-zA-Z_][a-zA-Z0-9_]*)\\s*\\}\\}/g — so a template can't reference dotted/hyphenated/\nleading-digit keys (`{{a.b}}`, `{{../x}}`, prototype-walking or \n[…]\n{{1e}}') === [].\n\nVerified: both FAIL when the charset is loosened to `[a-zA-Z0-9_.-]+`, and PASS on the real code.\n\nTest-only; no behaviour change. Full create-agent-harness suite 383/383; tsc clean.",
"is_bot": false,
"headline": "test(renderer): pin the {{var}} injection-safety charset so a looseni…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T01:47:04Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "096b17317dd0ac7fa53ac6cb454bbb461d49ca05",
"body": "…verified (#4 HIGH-1) (#124)\n\nBeta feedback #4 (HIGH-1): witness verification is a guaranteed no-op — no kernel backend exposes\nwitnessVerify, so witness-client.ts's crypto-verify delegate always falls through to the shape-only\ndegraded path, which returned {valid:true}. publish.ts only threw on !va\n[…]\n-gated): wiring a real witnessVerify so a\nGENUINE signature can be cryptographically checked. Until then publishing a \"signed\" harness requires\nthe explicit --allow-unverified-witness acknowledgement.",
"is_bot": false,
"headline": "fix(publish): fail closed when a present witness signature was never …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T01:27:45Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6a4d43e90e81d91bc6ad6b361d7c950ac4b7b770",
"body": "…lently disarm the gate (#4) (#125)\n\nBeta feedback #4 (mutation finding): mcp-scan is the CI security gate — mcpScanCmd exits 1 iff a HIGH\nfinding is present. Mutation testing showed that downgrading the `allow-shell` rule HIGH→MEDIUM SURVIVES\nthe suite: the existing tests assert finding IDs + the a\n[…]\n real code.\n- each security-critical rule (no-policy, no-default-deny, wildcard-tool-perm) keeps its HIGH severity.\n\nTest-only; no behaviour change. Full create-agent-harness suite 363/363; tsc clean.",
"is_bot": false,
"headline": "test(mcp-scan): pin per-finding HIGH severity so a downgrade can't si…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T01:19:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "cb3f4c7aa86d40312742d9d8b5dc3ac1912d7274",
"body": "…repros (ADR-175 §63, #47) (#122)\n\nDownstream feedback (#47, from ruflo) flags Conformant repair's Goodhart trap: the agent's\nself-written reproduce_bug.py IS the contract, so it can fail for a friendlier/unrelated reason than\nthe ticket demands. test-critic already rejects error-phase failures (Imp\n[…]\ninvalid and the resolve path are unchanged).\n\n7 tests (node-script style; $0, no Docker/LLM) wired into the CI node-test list. ADR-175 §63 updated\nto record the signal + the deferred-gating rationale.",
"is_bot": false,
"headline": "feat(darwin/repro): non-gating symptom-binding signal for conformant …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T01:04:11Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "0d83205dded6587f933d18f39aff47f9000651a8",
"body": "…(#4 HIGH-2, #7) (#123)\n\nBeta feedback #4 (HIGH-2): the \"bundle is sanitised\" promise was value-blind. Five copies of the same\nrecursive redactor (diag.ts, export-config.ts, threat-model.ts, genome.ts, score.ts) keyed ONLY on\nobject-KEY names, so a secret-shaped VALUE in a field NOT named secret/tok\n[…]\nkept, and the raw secret appears nowhere in the bundle.\n20 new tests (redact.ts precision + key/value redaction + the diag-bundle leak integration); full\ncreate-agent-harness suite 381/381; tsc clean.",
"is_bot": false,
"headline": "fix(redact): value-aware secret redaction across all output surfaces …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T01:02:31Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "094e3fd5abd389dc39d3197227bb9bd199981a35",
"body": "… unavailable (#22) (#121)\n\n`harness diag` reported the kernel VERSION but never WHICH backend actually\nanswers (native/wasm/js) — so an operator who set METAHARNESS_KERNEL_BACKEND=wasm\n(or expected native) got the silent js floor with no signal, the exact symptom\nin #22 (\"doctor → still reports js,\n[…]\nh a distinct\n\"(requested backend unavailable)\" label.\n\n7 tests (surfacing + end-to-end resolution + the requested-vs-uninstalled label\ndistinction). Full create-agent-harness suite 361/361; tsc clean.",
"is_bot": false,
"headline": "fix(diag): surface the resolved kernel backend + why a faster tier is…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-07T00:08:58Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c4002cc2f0c9690c9a038420aa813ff637916a9f",
"body": "…eaches the wasm backend (#20) (#120)\n\nEvery generated harness silently ran the JS floor backend even though a wasm\nartifact was shipped, because the publish workflow built `wasm-pack build\ncrates/kernel-wasm --target bundler` while the runtime loader (kernel-js\nloadWasm()) `import('../pkg/ruflo_ker\n[…]\nasm`-requested load resolves the wasm backend —\na bundler artifact would not). kernel-js suite 45 pass; tsc clean.\n\nCloses #20\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(kernel): publish the wasm as the nodejs target so `npm install` r…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T23:08:02Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "38284c8c49509decf14ce991bf17965433449e96",
"body": "…istinct from metaharness score (#15) (#119)\n\nThe `metaharness` and `harness` CLIs both accept `score` but emit DIFFERENT,\npurpose-specific JSON: `metaharness score` is the 5-dimension harness-fit\nscorecard (repo-scorecard, numeric `schema: 1`); `harness score` is the flat\nruntime-readiness badge se\n[…]\n not the numeric 1 (detectable vs\nmetaharness). score.ts is kernel-free so the suite runs $0. 15 score tests pass.\n\nCloses #15\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(score): harness score badge JSON carries a schema discriminator d…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T22:20:30Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "b601da5171e91f36f44b53024910aeb575fad577",
"body": "…antom `memory`/`route` CLI commands (#48) (#117)\n\nThe `minimal` template's CLAUDE.md advertised two subcommands its generated\nbin/cli.js never implements:\n | `<name> memory search <query>` | Semantic search across stored patterns |\n | `<name> route <task>` | Get the routing tier recommen\n[…]\nScan confirmed `minimal` was the\n only offender; failing-first verified (the test fails on the pre-fix template).\n\nCloses #48\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(create-agent-harness): minimal template docs no longer promise ph…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T22:07:51Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "de12ed8e868229e63ea7953753a39b94c75cb085",
"body": "… domain, $0 (ADR-237 §5) (#116)\n\nExecuted the full $0-local GSM8K flywheel run (qwen2.5-coder:7b, holdout=20,\nanchor=15, gens=4). Exact-match gold-scored; signed replay bundle committed.\n\nRESULT (honest, $0):\n- base accuracy 0.80 (16/20); 16 candidates over 4 gens → 0 promotions;\n milestone=false;\n[…]\nE-bench artifact. Reported plainly, no overclaim.\n\nArtifacts: proof-bundle-gsm8k.json (LIVE, replay PASS), analyze-gsm8k.json.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(evals-math): GSM8K live run evidence — HONEST NULL on a 2nd real…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T18:11:12Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "2db397c5e5a97586d3a8d6b029d43515b4386ab6",
"body": "…ain testbed (ADR-237) (#115)\n\nThe flywheel's central open question — does it compound on a REAL domain with a\nREAL model? — is gated on machine-hours for SWE-bench (ADR-236). Math isn't:\nGSM8K is a PUBLIC HF dataset (no token), gold answers are integers (#### N), and\ngrading is EXACT-MATCH — so a r\n[…]\ncall); 7b (~2-4s/call) does. The full experiment (~40 holdout, 6-8\ngens) is a one-command $0 launch — the immediate next step.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-math): LIVE GSM8K wiring — a fast, $0 real-compounding dom…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T17:51:20Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e6ab34964f5dfe170d91c95de0823657abc1ff86",
"body": "…lves solver behaviour, not just prompt prose (ADR-236 §6) (#114)\n\nThe D1 compounding-null (ADR-236 §4) is over-determined by the lever shape: the\npolicy levers (editPolicy/escalationPolicy/verifierPolicy) only APPEND system-prompt\nprose — the ADR-226 \"zero-marginal-advisor\" trap. A weak base model \n[…]\nuctural\nsplit + proposer menu); full swebench bench suite 139/139 (backward-compat: existing\neditPolicy-only tests unchanged).\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): structural capability lever for the D1 flywheel — evo…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T16:49:01Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "18293442385b718e406f9ce0da4b684c32097cde",
"body": "…udget gate (ADR-236 §5) (#113)\n\nD1's positive-compounding forward path (agentic solver + more generations/levers,\nADR-236 §4) was blocked on ONE thing: money. A hosted cheap model over dozens of\ninstances × generations × agentic steps is real $ + hours. This removes that gate\nfor a local OpenAI-com\n[…]\nqwen2.5-coder:32b): READY YES fully-local; footgun\n(hosted proposer vs local) correctly BLOCKED; hosted no-key path unchanged.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): $0 LOCAL-endpoint path for the D1 run — removes the b…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T16:39:08Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "b7276ceb615b9b7627781d59104796982214dbaa",
"body": "…y (0.1.7) (#112)\n\nThe package was ESM-only (exports had only an `import` condition), so a\nCommonJS consumer doing `require('@metaharness/flywheel')` hit\nERR_PACKAGE_PATH_NOT_EXPORTED — surfaced when meta-llm (CJS) became the first\nservice to depend on it (host-compat matrix failure). Publishing did\n[…]\nate fingerprint identical (b730019b),\n25/25 tests green. meetsPromotionRule source UNCHANGED (packaging only). 0.1.6 -> 0.1.7.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): dual CJS+ESM build so CommonJS consumers work nativel…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T15:01:10Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "bc64e3a6e8067c066c6ea091dd00aaf5be9fbf5b",
"body": "…unding null + anti-Goodhart save) (#111)\n\nThe user-authorized agentic positive-path run completed on REAL SWE-bench-Lite\n(official Docker harness gold-scoring), $0.0086 spend under the $15 cap.\n\nRESULT (honest):\n- Agentic base resolved 3/25 (12%) — a real 3x lift over the single-shot 1/25.\n- Flywhe\n[…]\n — NOT domain-proven compounding. Forward path:\nmore gens/levers + a larger holdout (3/25 leaves little headroom to compound).\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(swebench): D1-S5 — agentic run evidence (mechanism-proven, compo…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T14:46:23Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "fe08b82ef4fdb0b17941c3705966bc0e08154487",
"body": "…adapter (ADR-234, closes #109 SYNTHETIC-proven) (#110)\n\nCloses the unproven end-to-end piece of ADR-234 by mirroring evals-hle's\nadapter shape (typed genome → evaluator → composite gate → $0 synthetic\ntest) applied to the ruvllm SONA/MicroLoRA/EWC++ serving policy instead\nof an answer-production po\n[…]\nck.mjs's independently-\nversioned package set. Updates ADR-234 to Accepted-with-evidence and\ndocs/adrs/INDEX.md's summary row.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(ruvllm): @metaharness/evals-servedmodel — served-model flywheel …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T14:32:25Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "a5e3a14b6daea6ce1265d722f06e36fad579b880",
"body": "…mpletion (#108)\n\nA completed D1 run printed the lift curve but not WHICH policy levers earned\npromotions or at what cost. Wire the flywheel analyze (#104/#107) into the run:\nafter the lift curve, print formatAnalysis(analyzeBundle(bundle)) and write a\ndurable machine-readable analyze-swebench.json \n[…]\nomotions, per-lever avgΔ) + artifact written.\nNOTE: the workspace @metaharness/flywheel must be built before a run (standard).\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): auto-emit F-P2 analyze report + artifact on D1 run co…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T13:24:54Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "16476007fac94c0a72dcfc591c376e9248a53d28",
"body": "…pted-task) (0.1.6) (#107)\n\nExtends the 'analyze' CLI (#104): beyond WHICH levers promote, report at what\nCOST — using the sealed candidateScore.costPerWin already on every commit\n(ADR-235). This is the 'cost-per-accepted-task' dimension.\n\n- TargetStat gains avgCostPerWinPromoted + bestCostPerWin (p\n[…]\ns; root→current cost trajectory + direction). Smoke-verified\non the real D1 honest-null bundle (0 promotions → cpw shows '—').\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): analyze reports cost-per-win economics (cost-per-acce…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T13:22:41Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "1b003472c54cc941991770f154e0423c9d7adcb1",
"body": "…105)\n\nThe grader defaulted the swebench gold-scorer venv to /tmp/swebench-venv — but\n/tmp is wiped on reboot, so the harness silently vanished between sessions and a\nD1 run could not gold-score (import swebench → ModuleNotFoundError). This is the\nONLY gold-scorer; a missing venv must never be mista\n[…]\nts: 2/2 ($0 — missing-venv throws with setup\ninstructions; present-venv constructs). No gold-scoring here; nothing fabricated.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(swebench): durable harness venv path + loud missing-venv guard (#…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T13:00:53Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c317b6c6d147c3a76a90e98dc0d88113151bfbc0",
"body": "…an (#106)\n\nLaunching the budget-gated D1 flagship run should be a CONFIDENT one-command\naction. Add --plan/--dry-run: print the run shape (solver, holdout/anchor/gens/\ntargets, candidate count, solver-instance-run upper bound, HARD budget cap,\ncheckpoint+resume behavior) and a $0 pre-flight (Docker\n[…]\nEEN (Docker, ~/.cache/swebench-venv harness, suites, key),\n'READY TO LAUNCH: YES', exit 0. Nothing gold-scored — dry-run only.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): d1s4-live-run --plan — $0 dry-run pre-flight + run pl…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T12:58:07Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "9fdf9d4b95eb982a0b4172995880a839bc748d2f",
"body": "…ndle (#104)\n\nThe CLI had run|replay|graph but no way to answer 'which policy levers actually\nearn promotions?' — the F-P2 question that guides the next run's --targets.\n\nAdd `metaharness flywheel analyze <bundle>`: from a bundle's full candidate\nledger (all_commits), report per-lever attempts / pro\n[…]\n-promoted lever + per-lever stats; an\nhonest-null analyzes + formats cleanly). No engine change; meetsPromotionRule\nuntouched.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): 'analyze' CLI — F-P2 mutation-effectiveness from a bu…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T12:45:17Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "68ce14ae69d8a8f5add59ce2ee9b7382204b6f46",
"body": "… checkpoint (0.1.5) (#103)\n\nCompletes the crash-recovery story from #102. Checkpointing kept a crashed\nmulti-hour run's PARTIAL evidence, but finishing it meant restarting from\ngeneration 1 (re-spending). Now a run can RESUME from its last checkpoint.\n\n- @metaharness/flywheel 0.1.4 → 0.1.5:\n - Gen\n[…]\n —\nsame commit ids, final policy, lift curve, milestone, replay PASS). tsc clean.\nADR-236 updated. $0 — no SWE-bench run here.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): resumable runs — continue a crashed run from its last…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T12:32:42Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "a561ae6254c38317c40165ac71f591041b7d7d68",
"body": "…ecoverable long runs (#102)\n\nA LIVE agentic D1 run spans HOURS; runFlywheelGenerations only wrote the replay\nbundle at the very end, so a crash/OOM/rate-limit lost the whole run (the prior\nsingle-shot run crashed at ~45min with nothing saved).\n\n- @metaharness/flywheel 0.1.3 → 0.1.4: FlywheelConfig \n[…]\n can't kill the run); acceptance test confirms the buildBundle DRY\nrefactor is behavior-identical; tsc clean. ADR-236 updated.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): per-generation onGeneration checkpoint hook — crash-r…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T09:42:32Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "28349be573532c0c260302cf463aa81f75fa894b",
"body": "… + --solver flag (D1 positive-path prep) (#101)\n\nD1-S5 closed as an honest null because the base single-shot solve.mjs has no\nheadroom. The forward path is the multi-shot agentic solver — but solve-agentic.mjs\ndid NOT read SWE_POLICY_SYSTEM, so the flywheel could not evolve it and an\n--solver agent\n[…]\nsion); all scripts node --check clean. ADR-236 updated\n(forward path wired). No real SWE-bench result produced here ($0 prep).\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): wire the flywheel policy seam into the agentic solver…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T07:21:15Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "172bf6725c4b0b2e2a9230b8231b22155bb737ec",
"body": "…dapter (multi-vertical scale-up) (#96)\n\n* feat(evals-toolcall): @metaharness/evals-toolcall — BFCL-style tool-call flywheel adapter\n\nNew benchmark-domain adapter for @metaharness/flywheel, mirroring the shipped\n@metaharness/evals-hle template exactly. Treats BFCL-style function-calling /\ntool-use a\n[…]\n\\S] → polynomial on many spaces); .trim() already\nhandles leading ws. 8/8 tests pass.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-toolcall): @metaharness/evals-toolcall — flywheel domain a…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T06:32:01Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "1d83028f522e4f5dd17c79c43e8ec93741624b06",
"body": "… ADR-235 gate re-execution (#100)\n\nD1-S5 evidence. Two budgeted LIVE SWE-bench-Lite runs (glm-5.2, deepseek-chat) BOTH\nresolved 1/25 with 0 compounding lift (milestone=false), real official-Docker gold-scored,\nreplay PASS. PROVEN: the flywheel mechanism runs end-to-end on REAL code-repair (no\nfabri\n[…]\nvals-hle\n7/7 (gateReExecutes asserted). Commits the LIVE proof-bundle-swebench.json as evidence.\nmeetsPromotionRule UNCHANGED.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(adr): ADR-236 — D1 SWE-bench domain run is an HONEST NULL + wire…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T06:28:45Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6bea767392448cf466601bb5e0eea7985ab1cff0",
"body": "…lti-vertical scale-up) (#97)\n\n* feat(evals-sql): @metaharness/evals-sql — text-to-SQL flywheel adapter (mirrors evals-hle)\n\nNew benchmark domain adapter for @metaharness/flywheel: text-to-SQL (Spider-style)\ntreated as a POLICY-EVOLUTION problem — freeze the model, evolve the harness policy\nthat lin\n[…]\n[1] is .trim()'d downstream, so dropping \\s* is behaviour-preserving. 7/7 tests pass.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-sql): @metaharness/evals-sql — flywheel domain adapter (mu…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T06:16:21Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "96edb2900a2d9eace38972883f0a90fd2bd5446d",
"body": "…multi-vertical scale-up) (#95)\n\n* feat(evals-math): @metaharness/evals-math — GSM8K math-word-problem flywheel adapter\n\nNew benchmark-domain adapter for @metaharness/flywheel, mirroring the shipped\n@metaharness/evals-hle template exactly. Treats GSM8K-style math word problems as\na POLICY-EVOLUTION \n[…]\n\\d{1,15}\n(answers never exceed that) → linear. Behavior unchanged for real fractions.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-math): @metaharness/evals-math — flywheel domain adapter (…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T05:57:04Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "79236792631b561098a9fea7f152bba017fe94b1",
"body": "…res (ADR-235 forward plan) (#99)\n\nRealizes ADR-235 §2: an external reviewer can now RE-RUN the promotion rule on each\ncommit's sealed scores and confirm the verdict reproduces — trust the gate re-run, not\nthe logged verdict (the PR#64 \"trust no logs\" discipline, applied to promotion decisions).\n\n- \n[…]\nrule→false, no rule→true);\n flywheel suite 19/19; healthcheck 8/8. @metaharness/flywheel 0.1.3. meetsPromotionRule UNCHANGED.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): verifyReplayBundle re-executes the gate on sealed sco…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T05:47:30Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6217ce6d8a16640db3f77b2abbc2594688ffc982",
"body": "…on flywheel adapter (#98)\n\nNew benchmark-domain adapter mirroring @metaharness/evals-hle: structured JSON\nextraction from text as a POLICY-EVOLUTION problem for @metaharness/flywheel.\n\n- typed, bounded genome (schema-strictness, verification-mode json-schema-validate,\n field normalization, confide\n[…]\ntim\n- wired into scripts/build-ordered.mjs (phase 2, beside evals-hle) and the\n scripts/healthcheck.mjs INDEPENDENT allowlist\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-extract): @metaharness/evals-extract — structured-extracti…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T05:36:37Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "20ae3ee32dd7681d895be5fdbd8e5a4bb751726b",
"body": "…e budget guard sees real spend (#94)\n\nThe D1-S4 live run reported spend=$0.0086 for 25 real agentic solves — because solve.mjs\nread cost from `j.usage?.cost`, but OpenRouter only returns `usage.cost` when the request\nsends `usage: {include: true}`, which the two chat bodies OMITTED. So solver cost \n[…]\nre-run (a stronger solver, cost-bounded) for real domain\nevidence — the prior 1/25 with glm-5.2 was an honest null, not proof.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(swebench): solver cost tracking — send usage:{include:true} so th…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T05:26:35Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "8a5a916ea6022d24cb1143d42671f74cc424820d",
"body": "…iscipline (ADR-235) (#93)\n\nverifyReplayBundle wrongly FAILED an HONEST-NULL run (0 promotions): the chain is\njust the immutable gen-0 root — a valid, signed, replayable result — but allPromoted\nrequired >=1 promotion (`promos.length > 0 && ...`). The intent (\"no rejected node\nsmuggled into the prom\n[…]\nn each LineageCommit and re-run meetsPromotionRule to assert the verdict\nreproduces bit-for-bit. meetsPromotionRule UNCHANGED.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(flywheel): honest-null runs replay-PASS + re-executing-verifier d…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T05:16:45Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "1ffe33b45a84698b78ddc697a5c67472d13f2be3",
"body": "…he flywheel macro-loop (#91)\n\n* docs(adr): ADR-234 — ruvllm micro-loop (SONA/MicroLoRA/EWC++) under the flywheel macro-loop\n\nTwo-layer learning: ruvllm (per-request MicroLoRA weight adaptation + SONA + EWC++\nforgetting guard) as the micro-loop UNDER @metaharness/flywheel as the macro-loop\nstructura\n[…]\nly if >=3pp accuracy OR\n-25% cost/correct OR -50% regression, no anchor degradation).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(adr): ADR-234 — ruvllm micro-loop (SONA/MicroLoRA/EWC++) under t…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T02:48:24Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "62c72e60a5a6aba82035dc4237798437e1f9a566",
"body": "…) (#92)\n\nThe proposer complete() had no error handling; a transient 'fetch failed / write\nEPIPE' crashed the whole ~45min run (2026-07-06). Add fetchJSON with exp-backoff\nretry (EPIPE/ECONNRESET/5xx/429) + degrade complete() to '' (no mutation, safe) on\ngive-up. A prior commit of this fix did not make it into #90's merged tree — landing\nit on main here.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(swebench): D1-S4 runner fetch resilience (the fix that missed #90…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T02:47:31Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "934de686b34f712da1786dd729aa9e960ff5462f",
"body": "…un harness (dataset=Lite) (#90)\n\n* feat(swebench): D1-S4 scaffold — real Docker grader + budget-capped live-run harness\n\nswebench-grade.mjs wraps the OFFICIAL swebench harness (run_evaluation) as gradePredictions (the only\nthing that gold-scores). d1s4-live-run.mjs = the budgeted live flywheel run \n[…]\ncandidate\neval per generation. Keeps the authorized $15 / 25x2 pilot within envelope.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): D1-S4 — official Docker grader + budget-capped live-r…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T01:51:08Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "f43e210f67a7082c31c1ea176c2cb1d3ecdb592d",
"body": "…lywheel adapter (ADR-233) (#89)\n\n* feat(evals-hle): @metaharness/evals-hle — HLE as a policy-evolution flywheel adapter (ADR-233)\n\nTreat HLE (Humanity's Last Exam — closed-answer expert QA) as a policy-evolution\nproblem, not model training: freeze the models/evaluator/holdout, evolve only the\nharne\n[…]\nturns in 0ms; adapter tests stay 7/7; extraction\nbehavior preserved (ans42 / 42 / C).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(evals-hle): @metaharness/evals-hle — HLE as a policy-evolution f…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T01:36:24Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "49143d99c5a48eb8e65c3a94fa16899fd02f85d2",
"body": "… seam + $0 dry-run (#88)\n\nSelf-learning scale-loop iteration 3. Wires the REAL solver into the flywheel SWE-bench Evaluator so\nD1-S4 just swaps the endpoint.\n\n- solve.mjs: an OPERATING-POLICY seam — SWE_POLICY_SYSTEM env is appended to the base system prompt so\n the flywheel can evolve HOW the che\n[…]\nrness. Hard $-cap, cost-confirm first. meetsPromotionRule\nUNCHANGED; dataSource stays SYNTHETIC until the real harness scores.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): D1-S3 — real CLI runSolver (wraps solve.mjs) + policy…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T00:23:39Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "faa5b4fdae1432b6ceb64cce841cd64db54dca6d",
"body": "…or the flywheel domain run (#87)\n\nSelf-learning scale-loop iteration 2. Freezes the immutable, disjoint suites the D1-S4 lift curve is\nmeasured against.\n\n- freeze-swebench-suites.mjs: DETERMINISTIC hash-sorted disjoint split of full-300.json → a 40-instance\n HOLDOUT (optimized against) + a 15-inst\n[…]\nld-scored (or a --mock scorer)\n— validating the adapter end-to-end before the budgeted live run. meetsPromotionRule UNCHANGED.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): D1-S2 — frozen SWE-bench holdout (40) + anchor (15) f…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T00:09:54Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "47531b72c4835830dbeea469658c058470a16ac8",
"body": "…metaharness/flywheel (#86)\n\nSelf-learning scale-loop iteration 1. Bridges a flywheel POLICY + SWE-bench instances → a flywheel\nScore, so runFlywheelGenerations can compound operating-policy improvements on the REAL code-repair\ndomain (the \"reasoning proxy\" caveat killer). Everything SWE-bench-speci\n[…]\nlocks every real gain.\n\nNEXT (D1-S2): freeze a SWE-bench-Lite holdout + a separate anchor slice. meetsPromotionRule UNCHANGED.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(swebench): D1-S1 — SWE-bench code-repair Evaluator adapter for @…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-06T00:04:03Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "33f29f8e5fbedb0b12dcdf240e2cb69d0c44c3fc",
"body": "…I (run|replay|graph) (#84)\n\n* feat(flywheel): @metaharness/flywheel package + metaharness flywheel CLI (on 0.3.0 base)\n\nAdds the reusable, host/benchmark-agnostic promotion-loop engine and wires it into the umbrella CLI.\nRebased cleanly onto current main (the earlier branch was off a stale base).\n\n\n[…]\n own cadence (like darwin/redblue/weight-eft), so\nit joins the INDEPENDENT allowlist.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(flywheel): @metaharness/flywheel@0.1.1 + metaharness flywheel CL…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-05T23:28:56Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6e1902b6daff511d43b8132269018eafbf91e774",
"body": "… measured) (#82)\n\nThe ADR-240 G3 replay ran + gold-scored on 2026-07-04. Records the honest measured result,\ncorrecting the directive's methodological trap (an absolute >=6/24 bar would credit the advisor\nfor cand-6's executor lift — cand-6 D0 is already 8/24 with no advisor).\n\nIsolated the advisor\n[…]\n executor. Advisor track\nkilled; cand-6 serving-path injection (gated on this in ADR-235 §4.1) stays unwired. INDEX row added.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(adr): ADR-226 → Accepted/NULL — advisor track KILLED (G3 replay,…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-04T19:27:31Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "8c4c4f80e2d0c286547f84b869c2be1360f810d9",
"body": "…cost, optimize cost-per-accepted-task (ADR-232) (#81)\n\nOutput tokens can't be hidden (the model must generate them; in the claude -p loop\nany prose around the patch is billed then discarded), and Fable's output is 5x its\ninput ($50 vs $10/M). This adds a cost-aware output-mode decoder policy on the\n[…]\nDELED, not live. A 7-metric + 10-verbose-recover\nlive acceptance gate is specified and budget-gated/unrun (~$126 Fable spend).\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(cost): cost-aware output-mode decoder policy — cut Fable output …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-04T13:42:48Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "389029ee8128207785e62ce5bac81c4ed7c418d6",
"body": "…oop / localization_no_gold / best_of_n_selector_conformant from skip to enforceable (ADR-231) (#79)\n\n* feat(darwin): solver-trajectory forward-contract — flips no_gold_in_loop / localization_no_gold / best_of_n_selector_conformant from skip to enforceable (ADR-231)\n\nThe SWE-bench analog of ruflo #2\n[…]\ndn't touch ci.yml). Adds it to the existing $0 ADR-231 test step alongside\nsota-attest + nightly-sota-gate. Verified: exits 0.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(darwin): solver-trajectory forward-contract — flips no_gold_in_l…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T23:38:37Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "a25ef45bb61738c915e1d98c64ebae2e10faf037",
"body": "…exit non-zero (#73) (#78)\n\nbin.js awaited main() but discarded its resolved value, so EVERY metaharness\nsubcommand exited 0 regardless of failure — a failed command looked green in\nCI/scripts. PR #72 scope-fixed only `learn` (self-sets process.exitCode); this\nis the repo-wide fix.\n\n- bin.ts: set `p\n[…]\ns fail against pre-fix bin.js and pass after.\n\nNo command's behavior changes — only its exit code. Package version not bumped.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(cli): propagate main() exit code in bin.js — failed commands now …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T23:25:39Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "46ee0070552208ae76bc35799f7adcb05c0316d2",
"body": "…d integrity gate (#76)\n\n* docs(adr-231): provably-clean SOTA — signed submission-integrity attestation as a leaderboard gate\n\nAnswers UC Berkeley RDI (Apr 2026): all 8 major agent benchmarks were gamed to\n~98–100% without solving a task. Redefines a metaharness SOTA claim as the triple\n(score, cost\n[…]\n could break the rendered table / leave input\nbackslashes unescaped. Escape backslash FIRST, then pipe, then flatten newlines.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(ADR-231): provably-clean SOTA — working, enforced, Ed25519-signe…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T21:42:03Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "345b1abf833f547372168d1f0e4ba91fca3b338c",
"body": "…efault, cand-6 seed packaged) (#72)\n\n* feat(cli): metaharness learn — ADR-235 learning command ($0 dry-run default, cand-6 seed packaged)\n\nAdds a `learn` subcommand to the metaharness umbrella CLI that productizes the\nADR-235 managed-learning surface as a thin, repo-checkout-gated delegate to the\nG\n[…]\n9TuUo4eVSW7\n\n* chore: 0.2.8 → 0.3.0 (metaharness learn is a new command) + merge main\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(cli): metaharness learn — ADR-235 learning command ($0 dry-run d…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T16:53:49Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "d1a41098ccd0a83d21b9c48adaaf91256eccd340",
"body": "…a (pluggable evaluator) + cand-6 genome (#74)\n\n* feat(darwin): 0.8.0 — ship the GEPA engine as @metaharness/darwin/gepa + cand-6 genome\n\nPort the four portable GEPA modules (ADR-228) from bench/swebench/gepa/ to\ntyped TypeScript under src/gepa/, exposed as a new ./gepa subpath export\n(root export u\n[…]\nack intact.\n42/42 gepa tests pass unchanged; adversarial 100k-char input now instant.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(darwin): 0.8.0 — ship the GEPA engine as @metaharness/darwin/gep…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T16:30:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "0de26ad2b25741ef4e1a7d35192e48f83ebf635a",
"body": "…s-finops (#71)\n\n- metaharness CLI: bump 0.2.7 → 0.2.8 to publish the redblue + weight-eft\n umbrella subcommands (merged ~2026-06-25, never released). Standalone\n install proven: tarball installed in an isolated dir resolves\n @metaharness/redblue@0.1.4 + @metaharness/weight-eft@0.1.1 from the\n r\n[…]\n, coherent README, verifyProposal() proven by execution.\n Judged shippable; stays public and joins the publish list at 0.1.0.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "chore(pkg): 0.2.8 hygiene — publish prep, jujutsu drift reconcile, aw…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T16:27:26Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "b51803445d54215d33222e3f295bfece2df99792",
"body": "…breaker e2e timeout (#75)\n\n* fix(ci): raise cost-breaker e2e timeout to 30s — 5s default flakes on slow runners\n\nMain CI has been red since 865274d on 'completes and writes a winner under a tiny\nper-generation budget' timing out at vitest's 5000ms default on slow matrix runners\n(same run: ubuntu pa\n[…]\n preemptively: #71 syncs it to 0.2.0.\nVerified: node scripts/healthcheck.mjs exits 0.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "fix(ci): unbreak the Node matrix — path-guard false positives + cost-…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T15:46:35Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "aec01a2f4ef0dc557d673760a8419cb134dcd72d",
"body": "… cand-6 promotion (#70)\n\n* feat(gepa): concurrent + gateway-backed overnight training loop; land cand-6 promotion (ADR-228/235)\n\nClean rebuild of the overnight-training-loop work onto current main (old #69 was\ncut from a stale base and conflicted with the ADR-230 branch-memory work that\nhas since m\n[…]\ns to any log.\nVerified: dry-run output contains no env-var name; 10 learn tests pass.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(gepa): concurrent + gateway-backed overnight training loop; land…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T11:37:42Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "f63677f89c2f9343304783394734434e2e1f5889",
"body": "Fable-5 (anthropic/claude-fable-5, invoked directly via OpenRouter) surfaces\nfive novel, testable frontier optimizations for the MetaHarness control plane,\ngrounded in real state (cand-6 0.583->0.333, RLI ~16%, MCP-Atlas GLM-5.1 75.6%):\n\n- Q1 dense fitness shaping + residual-seeded mutation to cross\n[…]\nit. Fable's reasoning is attributed throughout;\neffect sizes/costs are Fable's estimates. Total Fable spend ~$1.62. Docs-only.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(research): Fable-guided frontier optimization opportunities (#68)",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T11:01:50Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "74e7bb8a384b35ec1ef98a1edc28fd83e6aac135",
"body": "…(primary) + tier-placement calibration (secondary) (#67)\n\n* docs(adr): RLI economic-work capability oracle for tier placement + vertical scoping (ADR-229)\n\nDeep-review Scale Labs' Remote Labor Index (RLI) and turn it into concrete\ntier + vertical-scoping decisions.\n\n- docs/research/rli-remote-labor\n[…]\nated to lead with the generative (RLI-Mini) framing.\n\nDocs only. No code, no secrets.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "docs(adr): RLI economic-workflow oracle — internal target generation …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-03T02:02:24Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "865274d6a90044d32039389fc83efb8f144527c1",
"body": "…ADR-230 pilot) — reconciled onto #64 (#65)\n\nReconciles the ADR-230 branch-memory deliverable onto the CURRENT main, whose #64 GEPA\nharness diverged from the vendored copy this work was originally built on: main's run-gepa.mjs\nhas NO `deriveLesson` callback / `liveDeriveLesson` and NO regression-rep\n[…]\niring records every evaluated candidate (incl. rejects) with\ncorrect parent lineage and promotes result.best on a holdout win.\n\n\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat(gepa): Agenticow branch-memory in the GEPA candidate lifecycle (…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T21:58:38Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c0768c1399cde9387bc5421eaf062bdfdbe8c1c3",
"body": "…gnitum-one/meta-llm (#66)\n\nPrevents the wrong-copy misdirect that hit ADR-229 (built here instead of the\ndeployed cognitum-one/meta-llm). The live gateway (flywheel, tier ladder,\ncache/midstream planes, Cloud Run rev 00071) is the separate cognitum-one/meta-llm\nrepo. This scaffold is frozen ~ADR-203; new gateway work here never reaches prod.",
"is_bot": false,
"headline": "chore: mark services/apicompletions as a DEPRECATED stale mirror → co…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T21:38:03Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "6b5090cdf8015a68e4ad12634610d24e832e8ca4",
"body": "…he cheap-executor policy (#64)\n\n* feat(apicompletions): scaffold Cognitum Fugu completions service skeleton (ADR-203)\n\nScaffold services/apicompletions/ — the OpenAI-compatible, metered, tiered\n/v1/chat/completions Cloud Run service from ADR-203 (rollout §7.4 step 3).\n\nStack matches the @cognitum-o\n[…]\nt holdout\ndelta (promotion reads holdout only). Pilot command + §9.5 gate documented.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01V84YJJK7RBF9TuUo4eVSW7",
"is_bot": false,
"headline": "feat: ADR-228 GEPA Distilled Executor Genome — offline evolution of t…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T21:32:03Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "16988b8bac014eae808dd6a90455e5064104eec9",
"body": "…nboarding & tier placement (#63)\n\nAdopts arXiv:2606.24020 (rank-2 matrix completion, 5-probe scorecard recovery,\nconformal confidence) as the MetaHarness model-onboarding layer. Novel extension:\ncolumns are (harness × slice) pairs, with the measured model×harness interaction\n(deepseek 35%-darwin/0%\n[…]\nr. Explicit non-goals: per-request escalation (ADR-225),\navailability gating (ADR-221). P0-P3 phases; leave-one-model-out backtest as the\nvalidation bar; ~10x cheaper placement ($60-125 → ≤$15/model).",
"is_bot": false,
"headline": "docs(adr): ADR-206 — BenchPress low-rank score prediction for model o…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T20:37:58Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "b89eab4ea8e7ba4d6da64f7eea63ee521182319b",
"body": "…tor (ADR-205) (#60)\n\n* feat(agentic): native OpenRouter tool-calling path for the ReAct loop (ADR-197)\n\nDiagnosed the fusion frontier arm's 0/8 empty-patch outcome by re-running live\nprobes against sonnet-5 on 3 hard-slice repos under the existing text-JSON\nprotocol: the model parses the protocol c\n[…]\n-as-models: deployable unit =\nmodel+prompt-protocol+tool-loop+verifier+retry-policy).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_01E27MdvFkLVcgbSqjzDCoVE",
"is_bot": false,
"headline": "feat(darwin): harness handoff — darwin→claude-p+Fable hard-tail actua…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T15:43:05Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c27b35a7a73d7cdf480ecadd258c78b0352ab805",
"body": "* fix(ci): rustfmt poker-darwin + ignore 2 unmaintained-crate advisories (bincode, paste)\n\nPre-existing Rust CI failures unrelated to any feature change:\n- cargo fmt --check: 5 spots in crates/poker-darwin (holdem.rs, convergence.rs)\n- cargo-deny advisories: RUSTSEC-2025-0141 (bincode) + RUSTSEC-202\n[…]\nt invocation (verified locally: EXIT=0, 311 deps scanned clean). Remove the dead\naudit.toml. deny.toml keeps its matching [advisories].ignore for cargo-deny.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>",
"is_bot": false,
"headline": "fix(ci): rustfmt + cargo-deny advisories (unblock PR #60) (#61)",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-07-02T15:38:15Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "7c5b3be0c7f71bfcd17b103d2189620a7bcb3375",
"body": "…ing, decoupled headroom flag, reservation leases)\n\nAddresses a peer review of the §5 Reserve-and-Commit budget-defense design. No\nbusiness-model / pricing / measured-vs-assumed claim changed — fixes are confined\nto the distributed budget mechanism (§5.2, new §5.5) and its risk register (§6).\n\nFix 1\n[…]\nion consistency edits\nin §5.3 + §8 test contract. Status unchanged: Proposed (rev 2).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr-204): rev 2 — peer review of Reserve-and-Commit (agent shard…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T21:12:22Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "53a6d45e727ee47232a8e288249508165d6fefe6",
"body": "…for Claude Code)\n\nAdd an Anthropic Messages API surface (POST /v1/messages + count_tokens helper)\nalongside the rev-1 OpenAI /v1/chat/completions surface, over the SAME\nauth/router/metering/budget/streaming core. Lets Claude Code + Anthropic-SDK\nclients point at Cognitum Fugu via ANTHROPIC_BASE_URL\n[…]\n).\n\nPurely additive: no OpenAI-protocol behaviour changes; all prior decisions stand.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr-203): rev 5 — dual-protocol surface (Anthropic Messages API …",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T21:09:53Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "fba4809b9b13060d7c03f6ee92fa98a953afa5db",
"body": "…udget-defense (companion to ADR-203)\n\nBusiness model + abuse-defense layer over ADR-203's tiered Completions API:\n- Two bundles: Sovereign Solopreneur ($99, low-tier-forced) and Agentic\n Autopilot ($249, low->mid->high escalation).\n- Tiered blended-cost model (C_blend = sum f_i*c_i); worked 75/20/\n[…]\ntion dangers cross-ref ADR-203 §5.1/§5.3.\n\nStatus: Proposed. Additive (new file only).\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr-204): unlimited-feeling flat-rate subscription economics + b…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T20:55:32Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "f13996d645bd6777575ef7827ffd0d62a1927a67",
"body": "…streaming\n\nAdditive only; does NOT change the rev-3 decision. Adds:\n- §3.5 \"Inflight streaming via ruvnet/midstream — optional, firewalled\":\n a better Option C' (inflight scan → mid-stream escalation without killing\n TTFT) that supersedes the rev-3 Option-C rejection ONLY when midstream is\n pres\n[…]\nagged unverified;\n integration written against midstream's actual API as a contract.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr-203): rev 4 — optional firewalled ruvnet/midstream inflight …",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T19:50:27Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "ccec7f73670e8c67c43e1b778abd7b9be8339ab0",
"body": "… drift, COUNT() cost\n\nThird peer-review round on Cognitum Fugu completions API:\n\n- Streaming escalation paradox (§3.3/§6.5/§3.4): post-gen τ escalation is\n incompatible with stream:true (tokens already sent). Decision: streams route\n once up front on the intrinsic input signal (can pick high dire\n[…]\nd Status to rev 3.\nOpen dependency unchanged: accountId on cognitum-one/api api_keys.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr-203): rev 3 — streaming-escalation paradox, tokenizer-family…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T19:46:11Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "181a6febf519c063ac90018afef52a39f1773e10",
"body": "…ream-truncation billing, scope-mismatch UX, pricing, τ\n\nMove ADR-203 (Cognitum Fugu completions API) Proposed→ready-for-Approved by\naddressing the five peer-review items:\n\n1. Firestore hot-spotting (§2.5/§5.3/§7.1/§7.4/§10): replace single-doc\n transactional sliding-window counters (~1 write/sec/\n[…]\n (§3.4). Remaining\nopen dependency unchanged: accountId on cognitum-one/api api_keys.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(ADR-203): address peer review (rev 2) — Firestore throughput, st…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T19:39:44Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "cba6d3cb986b8b182cfc4d75b6a0643bcc24599f",
"body": "… on MetaHarness\n\nProposed design: OpenAI-compatible /v1/chat/completions behind api.cognitum.one,\nauthed with the real cog_ API-key scheme (Firestore api_keys, SHA-256, scoped\npermissions), 3-tier difficulty/SLA routing via MetaHarness, metered with the\nagentbbs-gcp Firestore+Pub/Sub pattern. Tieri\n[…]\nacement\nPareto data; honest Fugu analog (no trained coordinator). Emulator-first, $0.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(adr): ADR-203 Cognitum Fugu — GCP metered tiered Completions API…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T19:31:55Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "dd1ac5727eaa83ad8641eafcf21add3328bad786",
"body": "…); cheap-tier ceiling is structural (no beyond-SOTA)\n\nTests the last non-orchestration lever: swap the cheap base model\ndeepseek-v4-pro -> glm-5.2, routing OFF, i2 harness fixed. Held-out 48-cell\ngrid (model{deepseek-v4-pro,glm-5.2} x language{py,ts,go,rust} x task{crud,cli}\nx 3 reps). Spend $4.73 \n[…]\nse-v4.csv, placement-analysis-v4.json,\nharvest4.py, analyze4.py. Posted to issue #38.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(retort): iteration-4 — stronger cheap model (glm-5.2 vs deepseek…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T16:29:12Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "eb7762053028cdd3ded64cf35f7dca2d40272763",
"body": "…tudio (relative Vite base) (#58)\n\nThe GitHub Pages site at https://ruvnet.github.io/metaharness/ rendered BLANK\nafter the repo rename: Vite baked asset URLs to the OLD base\n(/agent-harness-generator/assets/*.js|css), which 404 under the new\n/metaharness/ Pages base, so React never mounted.\n\nRoot ca\n[…]\n.. + ruvnet.github.io/... URLs, clone `cd`, and the\nREADME \"Repo:\" / SOTA \"Owner:\" repo-identity slugs updated to metaharness.\n\n\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "fix(pages): rename agent-harness-generator → metaharness; fix blank S…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T15:49:51Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "7a53e0d9fa31b88089c848b49b8d5d2d07e2d093",
"body": null,
"is_bot": false,
"headline": "Merge branch 'feat/retort-metaharness-runner'",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T14:10:44Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "33491fb7de1fe88d3e56fbb6a1aea63a0e8d9ddd",
"body": "…ss-runner",
"is_bot": false,
"headline": "Merge remote-tracking branch 'origin/main' into feat/retort-metaharne…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T14:10:12Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "b4b0bb0412251ddf6ed82610198e191390445775",
"body": "…unch\" confirmed (no beyond-SOTA)\n\nThe last orchestration lever. Added an opt-in INTRINSIC difficulty router to\ngreenfield-solve.mjs (--route-difficulty): a-priori (hard language + spec\nrequirement-density) + dynamic (token-burn / rewrite-churn / build-failures,\ngated by everRanOk for precision) sig\n[…]\nrouted-v3.csv, placement-analysis-v3.json, harvest3.py,\nanalyze3.py; PLACEMENT.md §7.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(retort): iteration-3 — task-difficulty-aware routing; \"no free l…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T14:09:48Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6a32b9332f15fcf3025eb8aee8b5e616cd5dd1f1",
"body": "…factor; beyond-SOTA not reached)\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf\n\n# Conflicts:\n#\tdocs/research/retort-placement/PLACEMENT.md",
"is_bot": false,
"headline": "Merge retort metaharness iteration-2 into main (timeout fix + memory …",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T11:30:14Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "95a150a895cbc91e706e4a8eeb7ef3e0d35ffb22",
"body": "…cap) + agenticow memory factor\n\nThe Phase-2 diagnosis isolated one genuine, non-overfitting lever: all metaharness\ntooling-fails were 12-min timeouts that crushed metaharness/cheap's pass-rate\n(0.62) despite high coverage. This iteration fixes it and tests memory, on a\nheld-out cheap-tier re-run (4\n[…]\nsis-v2.json,\nharvest2.py, analyze2.py; PLACEMENT.md §6. Posted to adrianco/retort#38.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(retort): iteration-2 — fix timeout lever (multi-action + 20-min …",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T11:29:17Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c84169b0507ee22feeeb48eeac8656bbe6cb381f",
"body": "… Pareto-optimal\n\nFirst real variance-attributed placement of the MetaHarness stack vs the\nclaude-code baseline inside Adrian Cockcroft's Retort DoE/ANOVA harness\n(adrianco/retort#38). 96-run grid (harness x model_tier x 4 langs x 2 tasks\nx 3 reps), Retort's own scorers + spec-gate judge reused unto\n[…]\nENT.md, pareto-frontier.png, results-combined.csv (96 rows),\nplacement-analysis.json.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(retort): Phase-2 Pareto placement + ANOVA — metaharness/cheap is…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T07:40:24Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "2b298f6410b22c8ed4a29f1329fff347a2873bd4",
"body": "… Pareto-optimal\n\nFirst real variance-attributed placement of the MetaHarness stack vs the\nclaude-code baseline inside Adrian Cockcroft's Retort DoE/ANOVA harness\n(adrianco/retort#38). 96-run grid (harness x model_tier x 4 langs x 2 tasks\nx 3 reps), Retort's own scorers + spec-gate judge reused unto\n[…]\nENT.md, pareto-frontier.png, results-combined.csv (96 rows),\nplacement-analysis.json.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "docs(retort): Phase-2 Pareto placement + ANOVA — metaharness/cheap is…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T07:40:13Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "4c3db7874b9e8d242806dec55f80ac69a084f060",
"body": "The Retort adapter for the darwin-mode agentic harness. Reuses the agentic-loop.mjs\nReAct core (parseAction / makeTools / stateHash / anti-thrash convo protocol) on a\nGREENFIELD \"implement this spec in language L\" task: empty workspace + TASK.md →\nthe model writes code, builds, runs its own tests, f\n[…]\npt.\nEmits one JSON telemetry line (tokens/cost/steps/model/escalated) for the runner.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(darwin-bench): retort metaharness greenfield solver",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T02:33:09Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "184116112d4a322680e0b9f91cfe6a86e96940c2",
"body": "…stone \"smarter orchestration\" test (FRAMES)\n\nCapstone experiment: does a learned, context-aware routing policy beat the best\nSTATIC policy on accuracy-per-dollar? Built a context-bucketed two-signal\nThompson-Sampling bandit (the @ruvector/rvf-solver PolicyKernel algorithm,\nreimplemented — the npm p\n[…]\npend: OpenRouter +$31.87 (authorized +$40; gate $2722, hard ceiling $2839 untouched).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(orchestration): learned adaptive router vs best static — the cap…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T00:04:18Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "4f7a77e24b9b42fe13f59cc6c7e0d27d1b55868c",
"body": "…stone \"smarter orchestration\" test (FRAMES)\n\nCapstone experiment: does a learned, context-aware routing policy beat the best\nSTATIC policy on accuracy-per-dollar? Built a context-bucketed two-signal\nThompson-Sampling bandit (the @ruvector/rvf-solver PolicyKernel algorithm,\nreimplemented — the npm p\n[…]\npend: OpenRouter +$31.87 (authorized +$40; gate $2722, hard ceiling $2839 untouched).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(orchestration): learned adaptive router vs best static — the cap…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-29T00:04:08Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "93da63a7e713c6b0a06d84707ec6c261dec0b1ef",
"body": "…ory + Darwin evolve is NULL on a cheap model\n\nEmpirically tests whether harness-evolve + agenticow branchable memory yields a\nmeasurable cognitive lift for deepseek-v4-pro on FRAMES (GAIA-class multi-hop QA,\nn=40, seed 42, reasoning OFF). New isolated harness under\npackages/darwin-mode/bench/cognit\n[…]\nort gate never tripped).\n\nReport: docs/research/cognition/COGNITION-EVOLVE-RESULTS.md\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(cognition): memory-as-cognition A/B/C — agenticow branchable mem…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T23:55:39Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "9b443b869c2592b51990c397b557171e8039e23b",
"body": "…ory + Darwin evolve is NULL on a cheap model\n\nEmpirically tests whether harness-evolve + agenticow branchable memory yields a\nmeasurable cognitive lift for deepseek-v4-pro on FRAMES (GAIA-class multi-hop QA,\nn=40, seed 42, reasoning OFF). New isolated harness under\npackages/darwin-mode/bench/cognit\n[…]\nort gate never tripped).\n\nReport: docs/research/cognition/COGNITION-EVOLVE-RESULTS.md\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(cognition): memory-as-cognition A/B/C — agenticow branchable mem…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T23:55:30Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "a7d4b8a04c4b82e919bb698f0fa714e436102851",
"body": "…R #619 (#57)\n\nAdds VECTOR-SEARCH-PROOF.md with real measured numbers from full SIFT-1M\n(1M vectors × 128-dim, AMD Ryzen 9 9950X, M=16 efC=200 single-threaded).\n\nKey findings (honest, not projected):\n\n * Build throughput: before 849 s / 1177 vec/s → after 774 s / 1292 vec/s\n \n[…]\nepo's OPTIMAL-RUVECTOR-APPLICATIONS.md (that was measured on a\n100K synthetic dataset where the working set fits in L3 cache).\n\n\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "research(vector-search): SIFT-1M measured before/after for ruvector P…",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T21:14:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c4d77975e93a6475a89be9c7db076bac04703c62",
"body": "…R #619\n\nAdds VECTOR-SEARCH-PROOF.md with real measured numbers from full SIFT-1M\n(1M vectors × 128-dim, AMD Ryzen 9 9950X, M=16 efC=200 single-threaded).\n\nKey findings (honest, not projected):\n\n * Build throughput: before 849 s / 1177 vec/s → after 774 s / 1292 vec/s\n = +9.7\n[…]\nhat was measured on a\n100K synthetic dataset where the working set fits in L3 cache).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "research(vector-search): SIFT-1M measured before/after for ruvector P…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T21:11:55Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "40bfb03190a3a02691132af7791ba4ca159b06ec",
"body": "…(cheap models) — scaffolds backfire; only Self-Consistency lifts (not significant)\n\nQ: does a better reasoning loop raise cheap-model resolve, at what cost?\nAblation {base, Plan-and-Solve, Self-Consistency N=10, verifier-BoN N=10, PS+BoN}\n× {deepseek-v4-pro, glm-5.2} on FRAMES n=50 seed 42, 12-step\n[…]\nns only); strict GAIA-EM scored offline. Evidence: runs/preds-*.jsonl + summary.json.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(scaffolding): agent-scaffolding intelligence ablation on FRAMES …",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T20:19:06Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "5adcb08b4bf9ced7b762cb42a42617e266f4f619",
"body": "…+ agenticow dual-state bridge (ADR-202) (#56)\n\nWraps agentic-jujutsu (lock-free jj op-log, QuantumDAG coordination,\nReasoningBank trajectories, ML-DSA signing) as a REMOVABLE MetaHarness\ncapability (ADR-150 principle: optional peer deps, graceful degradation,\nhonest probe()), and bridges it 1:1 to \n[…]\nhe jj 0.35.0 it bundles. Adapter uses `bookmark` with a\nbranchCreate fallback; upstream fix to be PR'd to ruvnet/agentic-flow.\n\n\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(jujutsu): add @metaharness/jujutsu — version-control-for-agents …",
"author_name": "rUv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T18:18:38Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "01dfa505bd73b0f8443a69460dcf840964febb2c",
"body": "…tions with proof. OPTIMAL: RVF COW snapshot/lineage (novel) + embedded GraphRAG substrate (niche). NULL: RAG-for-cheap-LLMs (4 ways, structural) + ANN speed (3-5x behind hnswlib). Un-refuted: confidence-routing + large-repo hybrid retrieval. Bug fixed PR #616.",
"is_bot": false,
"headline": "research(ruvector-applications): capstone synthesis — optimal applica…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T16:15:44Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "158955b19da87863bc0cea0a9ce81bcaec055f46",
"body": "…pin-image-reuse",
"is_bot": false,
"headline": "Merge remote-tracking branch 'origin/main' into claude/cve-bench-era-…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T16:15:44Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "98dda74156415ecb94401fbb958fa05f5ef924aa",
"body": "… paid micro-pilot\n\nCode-axis reformulation of H3 (leverage-map Priority-2; RepoGraph arXiv 2408.09504).\nJS-only parser (no Python) builds a file graph from import/test→target topology;\ntraversal uses real local @ruvector/graph-node v2.0.4 kHopNeighbors + JS PageRank.\n\n- Gate ($0, 5 repos): graphHit\n[…]\nh3-code-chart.mjs, Fig 10;\nupdates ADR-201 + REPORT.md + VECTOR-MEMORY-H3-RESULTS.md.\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(ADR-201): H3-code (SWE-bench) — code-axis graph retrieval gate +…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T16:13:54Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "dc3d1ea595ecf03ca6ba773db3575c5d6350258f",
"body": "…pin-image-reuse\n\n# Conflicts:\n#\tdocs/adrs/ADR-201-vector-memory-graphrag-cheap-model-lift.md",
"is_bot": false,
"headline": "Merge remote-tracking branch 'origin/main' into claude/cve-bench-era-…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T16:12:55Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "88323233182cf5aa8eb51b7f68926cc5bbf1ad67",
"body": "… paid micro-pilot\n\nThe FRAMES H3 was a structural null (Wikipedia+ONNX = dense cosine cluster →\ncosine-thresholded graph fully connected → kHop≡dense, graphHits=0). H3-code\ntests the opposite axis: build a code graph from import/test→target TOPOLOGY\n(not cosine) so traversal surfaces structurally-c\n[…]\n-MEMORY-H3-RESULTS.md (surgical, coordinated\nwith the concurrent router-pilot edits).\n\nCo-Authored-By: claude-flow <ruv@ruv.net>\nClaude-Session: https://claude.ai/code/session_019rVRYrRDKyxYK18kuVrDSf",
"is_bot": false,
"headline": "feat(ADR-201): H3-code (SWE-bench) — code-axis graph retrieval gate +…",
"author_name": "ruv",
"author_login": "ruvnet",
"committed_at": "2026-06-28T16:11:24Z",
"body_truncated": true,
"is_coding_agent": false
}
],
"releases_count": 1,
"commits_last_year": 880,
"latest_release_at": "2026-06-14T16:48:00Z",
"latest_release_tag": "v0.1.3",
"releases_from_tags": false,
"days_since_last_push": 6,
"active_weeks_last_year": 6,
"days_since_latest_release": 39,
"mean_days_between_releases": null
},
"community": {
"has_readme": true,
"has_license": true,
"has_description": true,
"has_contributing": true,
"health_percentage": 85,
"has_issue_template": false,
"has_code_of_conduct": true,
"has_pull_request_template": false
},
"ecosystem": {
"packages": [
{
"name": "@metaharness/codex",
"exists": true,
"license": "MIT",
"keywords": [
"metaharness",
"agent-harness",
"agent-harness-generator",
"codex",
"host",
"ai-agent",
"mcp",
"npx",
"scaffold"
],
"ecosystem": "npm",
"matches_repo": false,
"registry_url": "https://www.npmjs.com/package/@metaharness/codex",
"is_deprecated": false,
"latest_version": "0.1.0",
"repository_url": "https://github.com/ruvnet/agent-harness-generator",
"versions_count": 1,
"total_downloads": null,
"dependents_count": null,
"deprecation_note": null,
"maintainers_count": 1,
"monthly_downloads": 35,
"first_published_at": "2026-06-14T20:23:21.740000Z",
"latest_published_at": "2026-06-14T20:23:22.054000Z",
"latest_version_yanked": null,
"days_since_latest_publish": 39
},
{
"name": "@metaharness/coding",
"exists": true,
"license": "MIT",
"keywords": [
"metaharness",
"agent-harness",
"agent-harness-generator",
"coding",
"vertical",
"claude-code",
"ai-agent",
"mcp",
"npx",
"scaffold"
],
"ecosystem": "npm",
"matches_repo": false,
"registry_url": "https://www.npmjs.com/package/@metaharness/coding",
"is_deprecated": false,
"latest_version": "0.1.0",
"repository_url": "https://github.com/ruvnet/agent-harness-generator",
"versions_count": 1,
"total_downloads": null,
"dependents_count": null,
"deprecation_note": null,
"maintainers_count": 1,
"monthly_downloads": 39,
"first_published_at": "2026-06-14T20:23:29.630000Z",
"latest_published_at": "2026-06-14T20:23:29.941000Z",
"latest_version_yanked": null,
"days_since_latest_publish": 39
},
{
"name": "@metaharness/devops",
"exists": true,
"license": "MIT",
"keywords": [
"metaharness",
"agent-harness",
"agent-harness-generator",
"devops",
"vertical",
"claude-code",
"ai-agent",
"mcp",
"npx",
"scaffold"
],
"ecosystem": "npm",
"matches_repo": false,
"registry_url": "https://www.npmjs.com/package/@metaharness/devops",
"is_deprecated": false,
"latest_version": "0.1.1",
"repository_url": "https://github.com/ruvnet/agent-harness-generator",
"versions_count": 2,
"total_downloads": null,
"dependents_count": null,
"deprecation_note": null,
"maintainers_count": 1,
"monthly_downloads": 394,
"first_published_at": "2026-06-14T20:23:44.353000Z",
"latest_published_at": "2026-06-15T01:24:35.030000Z",
"latest_version_yanked": null,
"days_since_latest_publish": 38
}
]
},
"popularity": {
"forks": 58,
"stars": 510,
"watchers": 1,
"fork_history": {
"days": [
{
"date": "2026-06-13",
"count": 1
},
{
"date": "2026-06-14",
"count": 1
},
{
"date": "2026-06-15",
"count": 3
},
{
"date": "2026-06-16",
"count": 3
},
{
"date": "2026-06-17",
"count": 4
},
{
"date": "2026-06-18",
"count": 4
},
{
"date": "2026-06-19",
"count": 2
},
{
"date": "2026-06-20",
"count": 3
},
{
"date": "2026-06-21",
"count": 2
},
{
"date": "2026-06-22",
"count": 2
},
{
"date": "2026-06-23",
"count": 1
},
{
"date": "2026-06-24",
"count": 1
},
{
"date": "2026-06-25",
"count": 3
},
{
"date": "2026-06-27",
"count": 1
},
{
"date": "2026-06-28",
"count": 2
},
{
"date": "2026-07-04",
"count": 1
},
{
"date": "2026-07-05",
"count": 5
},
{
"date": "2026-07-06",
"count": 6
},
{
"date": "2026-07-07",
"count": 3
},
{
"date": "2026-07-09",
"count": 1
},
{
"date": "2026-07-10",
"count": 1
},
{
"date": "2026-07-12",
"count": 1
},
{
"date": "2026-07-14",
"count": 1
},
{
"date": "2026-07-16",
"count": 1
},
{
"date": "2026-07-19",
"count": 1
},
{
"date": "2026-07-21",
"count": 2
},
{
"date": "2026-07-22",
"count": 2
}
],
"complete": true,
"collected": 58,
"total_forks": 58
},
"star_history": null,
"open_issues_and_prs": 23
},
"ai_readiness": {
"has_nix": false,
"example_dirs": [
"examples"
],
"has_llms_txt": false,
"has_dockerfile": true,
"has_mcp_signal": false,
"bootstrap_files": [],
"api_schema_files": [],
"has_devcontainer": false,
"typecheck_configs": [
"apps/web-ui/tsconfig.json",
"packages/agent-harness-generator-lib/tsconfig.json",
"packages/aws-finops/tsconfig.json",
"packages/bench/tsconfig.json",
"packages/create-agent-harness/tsconfig.json",
"packages/darwin-mode/tsconfig.json",
"packages/evals-extract/tsconfig.json",
"packages/evals-hle/tsconfig.json",
"packages/evals-math/tsconfig.json",
"packages/evals-servedmodel/tsconfig.json",
"packages/evals-sql/tsconfig.json",
"packages/evals-toolcall/tsconfig.json",
"packages/flywheel/tsconfig.json",
"packages/harness/tsconfig.json",
"packages/host-claude-code/tsconfig.json",
"packages/host-codex/tsconfig.json",
"packages/host-copilot/tsconfig.json",
"packages/host-github-actions/tsconfig.json",
"packages/host-hermes/tsconfig.json",
"packages/host-openclaw/tsconfig.json",
"packages/host-opencode/tsconfig.json",
"packages/host-pi-dev/tsconfig.json",
"packages/host-rvm/tsconfig.json",
"packages/jujutsu/tsconfig.json",
"packages/kernel-js/tsconfig.json",
"packages/projects/tsconfig.json",
"packages/redblue/tsconfig.json",
"packages/router/tsconfig.json",
"packages/sdk/tsconfig.json",
"packages/vertical-base/tsconfig.json",
"packages/vertical-trading/tsconfig.json",
"packages/weight-eft/tsconfig.json",
"packages/workspace-lens/tsconfig.json",
"packages/workspace-probe/tsconfig.json",
"services/apicompletions/tsconfig.json"
],
"toolchain_manifests": [
"Cargo.toml",
"crates/kernel-napi/Cargo.toml",
"crates/kernel-wasm/Cargo.toml",
"crates/kernel/Cargo.toml",
"crates/poker-darwin/Cargo.toml",
"crates/template-catalog/Cargo.toml"
],
"largest_source_bytes": 57407,
"source_files_sampled": 1041,
"oversized_source_files": 0,
"agent_instruction_files": [],
"agent_instruction_max_bytes": null
},
"dependencies": {
"manifests": [
"Cargo.toml",
"package.json"
],
"advisories": {
"error": null,
"scope": null,
"source": null,
"findings": [],
"collected": false,
"malicious": [],
"truncated": false,
"by_severity": {},
"advisory_count": 0,
"affected_count": 0,
"assessed_count": 0,
"malicious_count": 0,
"assessed_package": null,
"unassessed_count": 0,
"direct_affected_count": 0
},
"ecosystems": [
"crates",
"npm"
],
"dependencies": [
{
"name": "@huggingface/transformers",
"manifest": "apps/web-ui/package.json",
"ecosystem": "npm",
"version_constraint": "^4.2.0"
},
{
"name": "jszip",
"manifest": "apps/web-ui/package.json",
"ecosystem": "npm",
"version_constraint": "^3.10.1"
},
{
"name": "lucide-react",
"manifest": "apps/web-ui/package.json",
"ecosystem": "npm",
"version_constraint": "^0.456.0"
},
{
"name": "react",
"manifest": "apps/web-ui/package.json",
"ecosystem": "npm",
"version_constraint": "^18.3.1"
},
{
"name": "react-dom",
"manifest": "apps/web-ui/package.json",
"ecosystem": "npm",
"version_constraint": "^18.3.1"
},
{
"name": "ruflo-kernel",
"manifest": "crates/kernel-napi/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "napi",
"manifest": "crates/kernel-napi/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "napi-derive",
"manifest": "crates/kernel-napi/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde",
"manifest": "crates/kernel-napi/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde_json",
"manifest": "crates/kernel-napi/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "ruflo-kernel",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde_json",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "wasm-bindgen",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde-wasm-bindgen",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "js-sys",
"manifest": "crates/kernel-wasm/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde_json",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "thiserror",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "ed25519-dalek",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "2.1"
},
{
"name": "sha2",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "0.10"
},
{
"name": "hex",
"manifest": "crates/kernel/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "0.4"
},
{
"name": "serde",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde_json",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "thiserror",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "ruvector-core",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "0.1.31"
},
{
"name": "candle-core",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "0.8"
},
{
"name": "candle-nn",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "0.8"
},
{
"name": "rs_poker",
"manifest": "crates/poker-darwin/Cargo.toml",
"ecosystem": "crates",
"version_constraint": "5"
},
{
"name": "serde",
"manifest": "crates/template-catalog/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
},
{
"name": "serde_json",
"manifest": "crates/template-catalog/Cargo.toml",
"ecosystem": "crates",
"version_constraint": null
}
],
"all_dependencies": {
"error": "GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository",
"source": null,
"packages": [],
"collected": false,
"truncated": false,
"total_count": null,
"direct_count": null,
"indirect_count": null
}
},
"maintainership": {
"issues": {
"open_prs": 4,
"merged_prs": 97,
"open_issues": 19,
"closed_ratio": 0.5,
"closed_issues": 19,
"closed_unmerged_prs": 7
},
"bus_factor": 1,
"bot_contributors": 0,
"top_contributors": [
{
"type": "User",
"login": "ruvnet",
"commits": 830,
"avatar_url": "https://avatars.githubusercontent.com/u/2934394?v=4"
},
{
"type": "User",
"login": "claude",
"commits": 49,
"avatar_url": "https://avatars.githubusercontent.com/u/81847?v=4"
},
{
"type": "User",
"login": "Thibaultjaigu",
"commits": 1,
"avatar_url": "https://avatars.githubusercontent.com/u/84420566?v=4"
}
],
"contributors_sampled": 3,
"top_contributor_share": 0.943
},
"quality_signals": {
"has_ci": true,
"has_tests": true,
"ci_workflows": [
"ci.yml",
"draco.yml",
"examples-packages-smoke.yml",
"pages-monitor.yml",
"pages.yml",
"publish.yml",
"published-smoke.yml",
"real-tools.yml",
"security.yml"
],
"has_docs_dir": true,
"linter_configs": [],
"has_editorconfig": true,
"has_linter_config": false,
"has_precommit_config": false
},
"security_signals": {
"lockfiles": [
"package-lock.json"
],
"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": "28 out of 28 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": 0,
"reason": "Found 0/28 approved changesets -- score normalized to 0",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#code-review"
},
{
"name": "Contributors",
"score": 6,
"reason": "project has 2 contributing companies or organizations -- score normalized to 6",
"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": 0,
"reason": "project was created within the last 90 days. Please review its contents carefully",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#maintained"
},
{
"name": "Packaging",
"score": null,
"reason": "packaging workflow not 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": 9,
"reason": "SAST tool detected but not run on all commits",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#sast"
},
{
"name": "Security-Policy",
"score": 10,
"reason": "security policy file detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#security-policy"
},
{
"name": "Signed-Releases",
"score": null,
"reason": "no releases found",
"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": 0,
"reason": "11 existing vulnerabilities detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#vulnerabilities"
}
],
"commit": "072b95c0a74610de008dca5473343a81619cef20",
"ran_at": "2026-07-23T23:02:29Z",
"aggregate_score": 4.4,
"scorecard_version": "v5.5.0"
},
"has_codeql_workflow": false,
"has_security_policy": true,
"has_dependabot_config": false
},
"contribution_flow": {
"collected": true,
"ci_last_run_at": "2026-07-23T14:51:21Z",
"oldest_open_prs": [
{
"number": 52,
"created_at": "2026-06-25T02:27:17Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 85,
"created_at": "2026-07-05T23:57:51Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 118,
"created_at": "2026-07-06T22:10:36Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 141,
"created_at": "2026-07-11T16:19:41Z",
"last_comment_at": null,
"last_comment_author": null
}
],
"last_merged_pr_at": "2026-07-17T22:47:21Z",
"ci_last_conclusion": "SUCCESS",
"oldest_open_issues": [
{
"number": 2,
"created_at": "2026-06-14T04:32:39Z",
"last_comment_at": "2026-06-14T04:42:20Z",
"last_comment_author": "ruvnet"
},
{
"number": 4,
"created_at": "2026-06-15T11:50:07Z",
"last_comment_at": "2026-07-07T00:40:32Z",
"last_comment_author": "ruvnet"
},
{
"number": 8,
"created_at": "2026-06-16T19:28:54Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 14,
"created_at": "2026-06-16T20:14:53Z",
"last_comment_at": "2026-06-17T00:03:05Z",
"last_comment_author": "ruvnet"
},
{
"number": 22,
"created_at": "2026-06-17T20:49:50Z",
"last_comment_at": "2026-07-06T22:20:50Z",
"last_comment_author": "ruvnet"
},
{
"number": 39,
"created_at": "2026-06-19T14:25:49Z",
"last_comment_at": "2026-06-23T05:03:20Z",
"last_comment_author": "ruvnet"
},
{
"number": 45,
"created_at": "2026-06-22T20:06:27Z",
"last_comment_at": "2026-06-26T01:02:45Z",
"last_comment_author": "ruvnet"
},
{
"number": 46,
"created_at": "2026-06-22T20:06:28Z",
"last_comment_at": "2026-06-26T01:02:44Z",
"last_comment_author": "ruvnet"
},
{
"number": 47,
"created_at": "2026-06-22T21:20:43Z",
"last_comment_at": "2026-07-07T00:21:16Z",
"last_comment_author": "ruvnet"
},
{
"number": 130,
"created_at": "2026-07-07T02:43:48Z",
"last_comment_at": "2026-07-07T04:17:56Z",
"last_comment_author": "ruvnet"
},
{
"number": 133,
"created_at": "2026-07-07T11:07:58Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 134,
"created_at": "2026-07-07T11:07:59Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 135,
"created_at": "2026-07-07T12:02:30Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 137,
"created_at": "2026-07-07T16:56:12Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 139,
"created_at": "2026-07-09T00:09:10Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 142,
"created_at": "2026-07-12T20:38:57Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 146,
"created_at": "2026-07-19T00:49:01Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 147,
"created_at": "2026-07-19T01:11:40Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 148,
"created_at": "2026-07-19T01:11:41Z",
"last_comment_at": null,
"last_comment_author": null
}
]
}
},
"config": {
"disabled_metrics": [],
"disabled_categories": [],
"disabled_components": {}
},
"source": {
"url": "https://github.com/ruvnet/metaharness",
"host": "github.com",
"name": "metaharness",
"owner": "ruvnet"
},
"metrics": {
"overall": {
"key": "overall",
"band": "moderate",
"name": "Overall health",
"note": null,
"notes": [],
"value": 63,
"inputs": {
"security": 44,
"vitality": 68,
"community": 71,
"governance": 49,
"engineering": 80
},
"components": []
},
"categories": [
{
"key": "vitality",
"band": "moderate",
"name": "Vitality",
"value": 68,
"weight": 0.22,
"metrics": [
{
"key": "development_activity",
"band": "moderate",
"name": "Development activity",
"note": null,
"notes": [],
"value": 58,
"inputs": {
"commits_last_year": 880,
"human_commit_share": 1,
"days_since_last_push": 6,
"active_weeks_last_year": 6
},
"components": [
{
"key": "push_recency",
"name": "Push recency",
"detail": "last push 6 days ago",
"points": 36,
"status": "met",
"details": [
{
"code": "push_recency",
"params": {
"days": 6
}
}
],
"max_points": 36
},
{
"key": "commit_cadence",
"name": "Commit cadence",
"detail": "6/52 weeks with commits",
"points": 4.2,
"status": "partial",
"details": [
{
"code": "commit_cadence_weeks",
"params": {
"weeks": 6
}
}
],
"max_points": 36
},
{
"key": "commit_volume",
"name": "Commit volume",
"detail": "880 commits in the last year",
"points": 18,
"status": "met",
"details": [
{
"code": "commits_last_year",
"params": {
"count": 880
}
}
],
"max_points": 18
},
{
"key": "openssf_scorecard_maintained",
"name": "OpenSSF Scorecard: Maintained",
"detail": "project was created within the last 90 days. Please review its contents carefully",
"points": 0,
"status": "missed",
"details": [],
"max_points": 10
}
]
},
{
"key": "release_discipline",
"band": "good",
"name": "Release discipline",
"note": "Excluded from scoring (no data or not applicable): OpenSSF Scorecard: Signed-Releases. Remaining weights renormalized.",
"notes": [
{
"code": "excluded_no_data",
"params": {
"components": [
"openssf_scorecard_signed_releases"
]
}
},
{
"code": "weights_renormalized",
"params": {}
}
],
"value": 84,
"inputs": {
"releases_count": 1,
"latest_release_tag": "v0.1.3",
"releases_from_tags": false,
"days_since_latest_release": 39,
"mean_days_between_releases": null
},
"components": [
{
"key": "ships_releases",
"name": "Ships releases",
"detail": "1 releases published",
"points": 27,
"status": "met",
"details": [
{
"code": "releases_published",
"params": {
"count": 1
}
}
],
"max_points": 27
},
{
"key": "release_recency",
"name": "Release recency",
"detail": "latest release 39 days ago",
"points": 36,
"status": "met",
"details": [
{
"code": "release_recency",
"params": {
"days": 39
}
}
],
"max_points": 36
},
{
"key": "release_cadence",
"name": "Release cadence",
"detail": "cadence unknown (single release)",
"points": 12.6,
"status": "partial",
"details": [
{
"code": "release_cadence_unknown",
"params": {}
}
],
"max_points": 27
},
{
"key": "openssf_scorecard_signed_releases",
"name": "OpenSSF Scorecard: Signed-Releases",
"detail": "no releases found",
"points": 0,
"status": "excluded",
"details": [
{
"code": "no_data",
"params": {}
}
],
"max_points": 10
}
]
},
{
"key": "abandonment",
"band": "excellent",
"name": "Abandonment",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"cap": null,
"state": "unverified",
"guards": [],
"signals": [],
"red_flag": false,
"multiplier_pct": 100,
"declared_reason": null,
"unverified_reason": "repository_too_young",
"unanswered_open_prs": null,
"unanswered_open_issues": null,
"days_since_last_merged_pr": null,
"days_since_last_human_commit": null,
"days_since_last_human_commit_is_floor": false
},
"components": [
{
"key": "project_is_still_maintained",
"name": "Project is still maintained",
"detail": "maintenance record not established from the collected data",
"points": 100,
"status": "met",
"details": [
{
"code": "abandonment_unverified",
"params": {}
}
],
"max_points": 100
}
]
}
],
"description": "Is the project alive — is code being written and are releases shipping?"
},
{
"key": "community",
"band": "good",
"name": "Community & Adoption",
"value": 71,
"weight": 0.18,
"metrics": [
{
"key": "popularity",
"band": "moderate",
"name": "Popularity & adoption",
"note": null,
"notes": [],
"value": 58,
"inputs": {
"forks": 58,
"stars": 510,
"watchers": 1,
"growth_state": "unverified",
"growth_factor_pct": 100,
"growth_unverified_reason": "no_history"
},
"components": [
{
"key": "stars",
"name": "Stars",
"detail": "510 stars",
"points": 43.9,
"status": "partial",
"details": [
{
"code": "stars",
"params": {
"count": 510
}
}
],
"max_points": 60
},
{
"key": "forks",
"name": "Forks",
"detail": "58 forks",
"points": 14.6,
"status": "partial",
"details": [
{
"code": "forks",
"params": {
"count": 58
}
}
],
"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": "excellent",
"name": "Community health",
"note": null,
"notes": [],
"value": 85,
"inputs": {
"has_readme": true,
"has_license": true,
"has_contributing": true,
"has_issue_template": false,
"has_code_of_conduct": true,
"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": 13.5,
"status": "met",
"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
}
]
}
],
"description": "Does the project have users, downloads, attention, and a welcoming setup for contributors?"
},
{
"key": "governance",
"band": "at_risk",
"name": "Sustainability & Governance",
"value": 49,
"weight": 0.24,
"metrics": [
{
"key": "maintainer_resilience",
"band": "critical",
"name": "Maintainer resilience (bus factor)",
"note": null,
"notes": [],
"value": 20,
"inputs": {
"bus_factor": 1,
"contributors_sampled": 3,
"top_contributor_share": 0.943
},
"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 94% of commits",
"points": 1.3,
"status": "partial",
"details": [
{
"code": "top_contributor_share",
"params": {
"share": 94
}
}
],
"max_points": 22.5
},
{
"key": "contributor_breadth",
"name": "Contributor breadth",
"detail": "3 contributors",
"points": 4.1,
"status": "partial",
"details": [
{
"code": "contributors_sampled",
"params": {
"count": 3
}
}
],
"max_points": 13.5
},
{
"key": "openssf_scorecard_contributors",
"name": "OpenSSF Scorecard: Contributors",
"detail": "project has 2 contributing companies or organizations -- score normalized to 6",
"points": 6,
"status": "partial",
"details": [],
"max_points": 10
}
]
},
{
"key": "responsiveness",
"band": "moderate",
"name": "Issue & PR responsiveness",
"note": null,
"notes": [],
"value": 59,
"inputs": {
"merged_prs": 97,
"open_issues": 19,
"closed_issues": 19,
"issue_closed_ratio": 0.5,
"closed_unmerged_prs": 7
},
"components": [
{
"key": "issue_resolution",
"name": "Issue resolution",
"detail": "50% of issues closed",
"points": 23.4,
"status": "partial",
"details": [
{
"code": "issues_closed_share",
"params": {
"share": 50
}
}
],
"max_points": 46.75
},
{
"key": "pr_acceptance",
"name": "PR acceptance",
"detail": "97/104 decided PRs merged",
"points": 35.7,
"status": "partial",
"details": [
{
"code": "decided_prs_merged",
"params": {
"merged": 97,
"decided": 104
}
}
],
"max_points": 38.25
},
{
"key": "openssf_scorecard_code_review",
"name": "OpenSSF Scorecard: Code-Review",
"detail": "Found 0/28 approved changesets -- score normalized to 0",
"points": 0,
"status": "missed",
"details": [],
"max_points": 15
}
]
},
{
"key": "stewardship",
"band": "good",
"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": 75,
"inputs": {
"followers": 10588,
"owner_type": "User",
"is_verified": null,
"owner_login": "ruvnet",
"public_repos": 197,
"account_age_days": 4983
},
"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": "10,588 followers of ruvnet",
"points": 25,
"status": "met",
"details": [
{
"code": "owner_followers",
"params": {
"count": 10588,
"login": "ruvnet"
}
}
],
"max_points": 25
},
{
"key": "track_record",
"name": "Track record",
"detail": "197 public repos, account ~13 yr old",
"points": 25,
"status": "met",
"details": [
{
"code": "public_repos",
"params": {
"count": 197
}
},
{
"code": "account_age_years",
"params": {
"years": 13
}
}
],
"max_points": 25
}
]
}
],
"description": "Will the project survive its people — bus factor, responsiveness, who backs it, and package upkeep?"
},
{
"key": "engineering",
"band": "good",
"name": "Engineering Quality",
"value": 80,
"weight": 0.2,
"metrics": [
{
"key": "engineering_practices",
"band": "good",
"name": "Engineering practices",
"note": null,
"notes": [],
"value": 74,
"inputs": {
"has_ci": true,
"has_tests": true,
"has_editorconfig": true,
"has_linter_config": false,
"has_precommit_config": false
},
"components": [
{
"key": "ci_workflows",
"name": "CI workflows",
"detail": "9 workflow(s)",
"points": 24,
"status": "met",
"details": [
{
"code": "ci_workflows",
"params": {
"count": 9
}
}
],
"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": 0,
"status": "missed",
"details": [],
"max_points": 16
},
{
"key": "pre_commit_hooks",
"name": "Pre-commit hooks",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 9.6
},
{
"key": "editorconfig",
"name": ".editorconfig",
"detail": null,
"points": 6.4,
"status": "met",
"details": [],
"max_points": 6.4
},
{
"key": "openssf_scorecard_ci_tests",
"name": "OpenSSF Scorecard: CI-Tests",
"detail": "28 out of 28 merged PRs checked by a CI test -- score normalized to 10",
"points": 20,
"status": "met",
"details": [],
"max_points": 20
}
]
},
{
"key": "documentation",
"band": "excellent",
"name": "Documentation",
"note": null,
"notes": [],
"value": 90,
"inputs": {
"topics": [
"agent-harness",
"agent-scaffolding",
"agentic-ai",
"agentic-framework",
"autonomous-agents",
"claude-code",
"codex",
"hermes-agent",
"mcp",
"mcp-server",
"multi-agent",
"openai-codex",
"pi-dev",
"scaffold",
"scaffolding",
"agent-harness-generator",
"create-agent-harness",
"openclaw",
"rvm",
"sandboxing"
],
"has_wiki": false,
"homepage": "https://Cognitum.One",
"has_readme": true,
"has_docs_dir": true,
"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": 25,
"status": "met",
"details": [],
"max_points": 25
},
{
"key": "documentation_homepage_site",
"name": "Documentation / homepage site",
"detail": "https://Cognitum.One",
"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": "20 topics",
"points": 10,
"status": "met",
"details": [
{
"code": "topics_count",
"params": {
"count": 20
}
}
],
"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": "at_risk",
"name": "Security",
"value": 44,
"weight": 0.16,
"metrics": [
{
"key": "security_posture",
"band": "at_risk",
"name": "Security posture",
"note": "Excluded from scoring (no data or not applicable): Packaging, Signed-Releases. Remaining weights renormalized.",
"notes": [
{
"code": "excluded_no_data",
"params": {
"components": [
"packaging",
"signed_releases"
]
}
},
{
"code": "weights_renormalized",
"params": {}
}
],
"value": 44,
"inputs": {
"source": "openssf_scorecard",
"checks_evaluated": 16,
"scorecard_version": "v5.5.0",
"checks_inconclusive": 2,
"scorecard_aggregate": 4.4
},
"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": "28 out of 28 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 0/28 approved changesets -- score normalized to 0",
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.5
},
{
"key": "contributors",
"name": "Contributors",
"detail": "project has 2 contributing companies or organizations -- score normalized to 6",
"points": 1.5,
"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": "project was created within the last 90 days. Please review its contents carefully",
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.5
},
{
"key": "packaging",
"name": "Packaging",
"detail": "packaging workflow not detected",
"points": 0,
"status": "excluded",
"details": [
{
"code": "no_data",
"params": {}
}
],
"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 detected but not run on all commits",
"points": 4.5,
"status": "partial",
"details": [],
"max_points": 5
},
{
"key": "security_policy",
"name": "Security-Policy",
"detail": "security policy file detected",
"points": 5,
"status": "met",
"details": [],
"max_points": 5
},
{
"key": "signed_releases",
"name": "Signed-Releases",
"detail": "no releases found",
"points": 0,
"status": "excluded",
"details": [
{
"code": "no_data",
"params": {}
}
],
"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": "11 existing vulnerabilities detected",
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.5
}
]
},
{
"key": "high_risk_jurisdiction_exposure",
"band": "excellent",
"name": "High-Risk Jurisdiction Exposure",
"note": "Only high-confidence self-published location evidence affects this multiplier. Ambiguous matches are review-only; country evidence is not proof of nationality, citizenship, legal registration, malicious intent, or sanctions status.",
"notes": [
{
"code": "jurisdiction_evidence_limits",
"params": {}
}
],
"value": 100,
"inputs": {
"meaning": "self-published location evidence; not nationality or citizenship",
"red_flag": false,
"exposures": [],
"policy_countries": [
"Russia",
"Iran",
"North Korea"
],
"review_only_matches": 0,
"assessed_self_published_locations": 2
},
"components": [
{
"key": "policy_exposure_multiplier",
"name": "Policy exposure multiplier",
"detail": "no confirmed policy-scope location match",
"points": 100,
"status": "met",
"details": [
{
"code": "jurisdiction_no_match",
"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": "moderate",
"name": "AI Readiness",
"value": 55,
"weight": 0,
"metrics": [
{
"key": "ai_agent_context",
"band": "at_risk",
"name": "Agent context & guidance",
"note": null,
"notes": [],
"value": 40,
"inputs": {
"has_llms_txt": false,
"legible_history_share": 0.97,
"agent_instruction_files": [],
"agent_instruction_max_bytes": null
},
"components": [
{
"key": "agent_instructions",
"name": "Agent instructions",
"detail": "no CLAUDE.md / AGENTS.md / editor rules",
"points": 0,
"status": "missed",
"details": [
{
"code": "no_agent_instructions",
"params": {}
}
],
"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": "97 of 100 human commits state their intent (structured subject or explanatory body)",
"points": 40,
"status": "met",
"details": [
{
"code": "legible_history",
"params": {
"legible": 97,
"sampled": 100
}
}
],
"max_points": 40
}
]
},
{
"key": "ai_verify_loop",
"band": "moderate",
"name": "Verify loop (build / test / typecheck)",
"note": null,
"notes": [],
"value": 56,
"inputs": {
"has_nix": false,
"has_tests": true,
"lockfiles": [
"package-lock.json"
],
"has_dockerfile": true,
"typed_language": true,
"bootstrap_files": [],
"has_devcontainer": false,
"has_linter_config": false,
"typecheck_configs": [
"apps/web-ui/tsconfig.json",
"packages/agent-harness-generator-lib/tsconfig.json",
"packages/aws-finops/tsconfig.json",
"packages/bench/tsconfig.json",
"packages/create-agent-harness/tsconfig.json",
"packages/darwin-mode/tsconfig.json",
"packages/evals-extract/tsconfig.json",
"packages/evals-hle/tsconfig.json",
"packages/evals-math/tsconfig.json",
"packages/evals-servedmodel/tsconfig.json",
"packages/evals-sql/tsconfig.json",
"packages/evals-toolcall/tsconfig.json",
"packages/flywheel/tsconfig.json",
"packages/harness/tsconfig.json",
"packages/host-claude-code/tsconfig.json",
"packages/host-codex/tsconfig.json",
"packages/host-copilot/tsconfig.json",
"packages/host-github-actions/tsconfig.json",
"packages/host-hermes/tsconfig.json",
"packages/host-openclaw/tsconfig.json",
"packages/host-opencode/tsconfig.json",
"packages/host-pi-dev/tsconfig.json",
"packages/host-rvm/tsconfig.json",
"packages/jujutsu/tsconfig.json",
"packages/kernel-js/tsconfig.json",
"packages/projects/tsconfig.json",
"packages/redblue/tsconfig.json",
"packages/router/tsconfig.json",
"packages/sdk/tsconfig.json",
"packages/vertical-base/tsconfig.json",
"packages/vertical-trading/tsconfig.json",
"packages/weight-eft/tsconfig.json",
"packages/workspace-lens/tsconfig.json",
"packages/workspace-probe/tsconfig.json",
"services/apicompletions/tsconfig.json"
],
"agent_commit_share": 0,
"toolchain_manifests": [
"Cargo.toml",
"crates/kernel-napi/Cargo.toml",
"crates/kernel-wasm/Cargo.toml",
"crates/kernel/Cargo.toml",
"crates/poker-darwin/Cargo.toml",
"crates/template-catalog/Cargo.toml"
],
"dependency_bot_commit_share": 0
},
"components": [
{
"key": "one_command_bootstrap",
"name": "One-command bootstrap",
"detail": "Cargo.toml, crates/kernel-napi/Cargo.toml, crates/kernel-wasm/Cargo.toml (toolchain convention, no task runner)",
"points": 12.6,
"status": "partial",
"details": [
{
"code": "toolchain_convention",
"params": {
"files": "Cargo.toml, crates/kernel-napi/Cargo.toml, crates/kernel-wasm/Cargo.toml"
}
}
],
"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": 0,
"status": "missed",
"details": [],
"max_points": 11
},
{
"key": "static_type_checking",
"name": "Static type checking",
"detail": "apps/web-ui/tsconfig.json, packages/agent-harness-generator-lib/tsconfig.json, packages/aws-finops/tsconfig.json, packages/bench/tsconfig.json, packages/create-agent-harness/tsconfig.json, packages/darwin-mode/tsconfig.json, packages/evals-extract/tsconfig.json, packages/evals-hle/tsconfig.json, packages/evals-math/tsconfig.json, packages/evals-servedmodel/tsconfig.json, packages/evals-sql/tsconfig.json, packages/evals-toolcall/tsconfig.json, packages/flywheel/tsconfig.json, packages/harness/tsconfig.json, packages/host-claude-code/tsconfig.json, packages/host-codex/tsconfig.json, packages/host-copilot/tsconfig.json, packages/host-github-actions/tsconfig.json, packages/host-hermes/tsconfig.json, packages/host-openclaw/tsconfig.json, packages/host-opencode/tsconfig.json, packages/host-pi-dev/tsconfig.json, packages/host-rvm/tsconfig.json, packages/jujutsu/tsconfig.json, packages/kernel-js/tsconfig.json, packages/projects/tsconfig.json, packages/redblue/tsconfig.json, packages/router/tsconfig.json, packages/sdk/tsconfig.json, packages/vertical-base/tsconfig.json, packages/vertical-trading/tsconfig.json, packages/weight-eft/tsconfig.json, packages/workspace-lens/tsconfig.json, packages/workspace-probe/tsconfig.json, services/apicompletions/tsconfig.json",
"points": 11,
"status": "met",
"details": [
{
"code": "file_list",
"params": {
"files": "apps/web-ui/tsconfig.json, packages/agent-harness-generator-lib/tsconfig.json, packages/aws-finops/tsconfig.json, packages/bench/tsconfig.json, packages/create-agent-harness/tsconfig.json, packages/darwin-mode/tsconfig.json, packages/evals-extract/tsconfig.json, packages/evals-hle/tsconfig.json, packages/evals-math/tsconfig.json, packages/evals-servedmodel/tsconfig.json, packages/evals-sql/tsconfig.json, packages/evals-toolcall/tsconfig.json, packages/flywheel/tsconfig.json, packages/harness/tsconfig.json, packages/host-claude-code/tsconfig.json, packages/host-codex/tsconfig.json, packages/host-copilot/tsconfig.json, packages/host-github-actions/tsconfig.json, packages/host-hermes/tsconfig.json, packages/host-openclaw/tsconfig.json, packages/host-opencode/tsconfig.json, packages/host-pi-dev/tsconfig.json, packages/host-rvm/tsconfig.json, packages/jujutsu/tsconfig.json, packages/kernel-js/tsconfig.json, packages/projects/tsconfig.json, packages/redblue/tsconfig.json, packages/router/tsconfig.json, packages/sdk/tsconfig.json, packages/vertical-base/tsconfig.json, packages/vertical-trading/tsconfig.json, packages/weight-eft/tsconfig.json, packages/workspace-lens/tsconfig.json, packages/workspace-probe/tsconfig.json, services/apicompletions/tsconfig.json"
}
}
],
"max_points": 11
},
{
"key": "reproducible_environment",
"name": "Reproducible environment",
"detail": "Dockerfile, lockfile",
"points": 10,
"status": "met",
"details": [
{
"code": "file_list",
"params": {
"files": "Dockerfile, lockfile"
}
}
],
"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": "no automated dependency updates observed",
"points": 0,
"status": "missed",
"details": [
{
"code": "no_dependency_automation",
"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": "excellent",
"name": "Code legibility for models",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"primary_language": "TypeScript",
"largest_source_bytes": 57407,
"source_files_sampled": 1041,
"oversized_source_files": 0
},
"components": [
{
"key": "type_checkable_code",
"name": "Type-checkable code",
"detail": "TypeScript (statically typed)",
"points": 45,
"status": "met",
"details": [
{
"code": "statically_typed_language",
"params": {
"language": "TypeScript"
}
}
],
"max_points": 45
},
{
"key": "manageable_file_sizes",
"name": "Manageable file sizes",
"detail": "0/1041 source files over 60KB",
"points": 55,
"status": "met",
"details": [
{
"code": "oversized_source_files",
"params": {
"kb": 60,
"sampled": 1041,
"oversized": 0
}
}
],
"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",
"Could not fetch crates package 'ruflo-kernel' from its registry",
"Could not fetch crates package 'ruflo-kernel-napi' from its registry",
"Could not fetch crates package 'ruflo-kernel-wasm' from its registry",
"Could not fetch crates package 'poker-darwin' from its registry",
"Could not fetch crates package 'template-catalog' from its registry",
"npm package '@metaharness/codex' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring",
"npm package '@metaharness/coding' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring",
"npm package '@metaharness/devops' points at a different repository (https://github.com/ruvnet/agent-harness-generator); excluded from ecosystem scoring",
"GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository"
],
"report_type": "repository",
"generated_at": "2026-07-23T23:02:57.746532Z",
"schema_version": "0.27.0",
"badge_url": "https://raw.githubusercontent.com/inspect-software/badges/main/v1/r/ruvnet/metaharness.svg",
"full_name": "ruvnet/metaharness",
"license_state": "standard",
"license_spdx": "MIT"
}
Registros relacionados
Repositorios inspeccionados que comparten etiquetas de catálogo o ecosistemas con ruvnet/metaharness. Leer en paralelo →
Las puntuaciones son señales, no garantías. Reflejan prácticas públicamente visibles en GitHub; no son una auditoría de código ni una garantía de seguridad.
Los datos ausentes se excluyen y los pesos se renormalizan; nunca se puntúan como cero. La metodología es versionada y abierta: métricas v1.13.0, esquema v0.27.0 — metodología completa · wiki de métricas.
