agent-tools — salud del software de código abierto — inspect.software

npm

67Buenoíndice de salud

SigRank MCP — AI operator evaluation and token cascade efficiency tools. The yield cascade + live leaderboard as MCP tools any agent can call. On-device token scanner + tabbed TUI dashboard. npm i -g sigrank

JavaScript★ 0↓ 12.3K/mes9 sept 2026

MIT9 sept 2026 · métricas 2.10.0

npm · Go

65Buenoíndice de salud

DeliciousBuding/dida-cli

Agent-friendly CLI for Dida365 / TickTick

Go · Shell★ 8↓ 115/mes9 sept 2026

MIT9 sept 2026 · métricas 2.10.0

npm

65Buenoíndice de salud

gmemmy/agent-scenario-loop

The scenario is the asset; runners are swappable. Define mobile app journeys once, run them through any agent, device, or profiler, and preserve stable evidence that tells agents and humans whether behavior improved, regressed, or can't be trusted yet.

TypeScript★ 3↓ 302/mes10 sept 2026

MIT10 sept 2026 · métricas 2.10.0

npm

65Buenoíndice de salud

koltyakov/browser-bridge

🤖 Browser Bridge for Agentic AI Development - browser extension, MCP, and skilled CLI

TypeScript · JavaScript★ 23↓ 647/mes17 sept 2026

MIT17 sept 2026 · métricas 2.10.0

63Moderadoíndice de salud

alvinunreal/lazyskills

mission control for agent skills

Go★ 24618 sept 2026

MIT18 sept 2026 · métricas 2.10.0

npm

63Moderadoíndice de salud

inkbox-ai/inkbox

SDK, skills, and examples for Inkbox: give AI agents a persistent identity with email, phone number, and internet address

TypeScript · Python · Rust★ 91↓ 3799/mes5 oct 2026

MIT5 oct 2026 · métricas 2.10.0

npm

63Moderadoíndice de salud

putervision/spc

High-performance zero-dependency static analysis tool enforcing NASA Power of Ten rules across 20 programming languages, plus AI agent skill, MCP server config, prompt template, and LLM model security auditing.

JavaScript★ 39↓ 5103/mes18 sept 2026

MIT18 sept 2026 · métricas 2.10.0

npm

62Moderadoíndice de salud

kitepon/Spotter

Audit agent for Claude Code that catches missed tool calls — separates spotting from doing. Your Claude answers, Spotter watches.

JavaScript★ 0↓ 4878/mes14 sept 2026

MIT14 sept 2026 · métricas 2.10.0

npm

62Moderadoíndice de salud

kitepon/Throughline

Cut ~90% of Claude Code's context usage while keeping nearly all the memory

JavaScript★ 4↓ 2627/mes13 sept 2026

MIT13 sept 2026 · métricas 2.10.0

Go · npm

62Moderadoíndice de salud

qyinm/gandalf

Infrastructure as Code for AI Agents. Declare MCP servers, skills, and hooks in gandalf.toml. Review before apply. Roll back safely.

Go★ 010 sept 2026

MIT10 sept 2026 · métricas 2.10.0

PyPI · Go · npm

62Moderadoíndice de salud

tonquoc0407/capybara

Terminal trace debugger for AI agents.

Go · Python★ 12↓ 18/mes20 oct 2026

MIT20 oct 2026 · métricas 2.10.0

npm

60Moderadoíndice de salud

1clawAI/1claw-mcp

MCP server for 1claw secrets vault — gives AI agents secure, just-in-time access to secrets

TypeScript★ 3↓ 6921/mes21 oct 2026

MIT21 oct 2026 · métricas 2.10.0

Go · PyPI

60Moderadoíndice de salud

Clawdlinux/agent-native-format

Token-minimal system views for AI agents. Translate Kubernetes and live state into compact Agent Native Format with Go tooling and governed execution primitives.

Go · Python★ 113 sept 2026

Licencia propia13 sept 2026 · métricas 2.10.0

PyPI

60Moderadoíndice de salud

GeckoVision/gecko-surf

The knowledge graph for APIs your agent can trust. Maps any API — even messy, paywalled, or on-chain ones — into a verified graph your agent traverses instead of guessing, and simulates any action to a receipt before money moves.

Python★ 6↓ 360/mes17 sept 2026

Apache-2.017 sept 2026 · métricas 2.10.0

npm · crates.io

60Moderadoíndice de salud

flowleap-ai/flowleap-cli

El repositorio no publica descripción.

Rust · JavaScript★ 1↓ 1041/mes16 sept 2026

MIT16 sept 2026 · métricas 2.10.0

npm

57Moderadoíndice de salud

funkyfunc/run-mcp

A thin MCP client for humans and agents — run any Model Context Protocol server, call its tools, read its stderr, and reconnect after each edit. REPL, headless CLI, or an MCP server your agent drives. No client config edits, no restarts.

TypeScript★ 0↓ 865/mes10 sept 2026

MIT10 sept 2026 · métricas 2.10.0

npm · Go

56Moderadoíndice de salud

amxv/webctx

fast multi-provider web search & full page scrape cli for agents

Go★ 1↓ 169/mes12 sept 2026

Apache-2.012 sept 2026 · métricas 2.10.0

54Moderadoíndice de salud

SohrabZ/x-browser-mcp

Read and post to X from your own logged-in browser session, exposed to local AI agents over MCP. No X API keys required.

Go★ 219 oct 2026

MIT19 oct 2026 · métricas 2.10.0

npm

54Moderadoíndice de salud

getpipher/vision

Capability-aware vision + paste extension for the pi coding agent. Delegates image analysis only when the active model is text-only; passes through natively for multimodal models (zero delegation).

TypeScript★ 13↓ 1588/mes16 sept 2026

MIT16 sept 2026 · métricas 2.10.0

53Moderadoíndice de salud

dandriscoll/muxray

Read live tmux panes as deterministic JSON for LLM coding agents — capture, diff, and Claude/Codex/Copilot status. Single Go binary, no network egress.

Go★ 014 sept 2026

MIT14 sept 2026 · métricas 2.10.0

PyPI

53Moderadoíndice de salud

eraycc/mcp-manager-gateway

MCP Manager Gateway - Make MCP discoverable and launchable on demand, just like Skills: the Agent only sees one stable entry point, and only when needed does it find and invoke the actual tools.

Python★ 6↓ 6845/mes20 oct 2026

Apache-2.020 oct 2026 · métricas 2.10.0

53Moderadoíndice de salud

z2z23n0/agent-capsule

Share a full Codex or Claude Code conversation as one link, then import it into a native local thread/session to continue.

Go · JavaScript★ 918 sept 2026

Apache-2.018 sept 2026 · métricas 2.10.0

PyPI · npm

51Moderadoíndice de salud

zencodeinc/hydracept-public

Execution infrastructure for AI agents. Run text, reasoning, image, audio, video, 3D, domains, DNS, and more through one capability API—with durable jobs, transparent pricing, receipts, budgets, and BYOK.

Python · HTML★ 0↓ 5402/mes21 oct 2026

MIT21 oct 2026 · métricas 2.10.0

50Moderadoíndice de salud

jhartum/redditrs

Go CLI for Reddit search & research: threads, subreddits, trends via web-session cookie — no API keys, no quota. Agent-friendly output.

Go · Shell★ 012 oct 2026

MIT12 oct 2026 · métricas 2.10.0

npm · PyPI · Go

48Débilíndice de salud

2s-io/sdk

Client API wrapper for 2s.io.

TypeScript · Python★ 8↓ 3890/mes22 oct 2026

MIT22 oct 2026 · métricas 2.10.0

npm

48Débilíndice de salud

jaein4722/Let-My-Agent-Sleep

Let AI agents start long-running jobs, stop polling, and resume the same session when the job finishes.

Shell · JavaScript★ 5↓ 125/mes10 sept 2026

MIT10 sept 2026 · métricas 2.10.0

npm

42Débilíndice de salud

hezhongtang/dsh-capability-optimizer

External-expert consultation for DeepSeek Harness: headless Claude Code CLI with role personas (advisor/reviewer/designer, extensible), replies as reference answers — thinking effort, model fallback, panels, settings workspace. · DSH 外部专家咨询：headless 调用 Claude Code CLI，角色人设（advisor/reviewer/designer，可扩展）回复即参考答案——推理等级、模型回退、并行会诊、设置工作区。

JavaScript★ 0↓ 1402/mes18 sept 2026

MIT18 sept 2026 · métricas 2.10.0

35Débilíndice de salud

dabito/clyph

Nerd Font glyph lookup CLI for shell scripts, status lines, and agent-friendly UI work. Search glyph names, print exact glyphs, return codepoints, and refresh a local offline catalog from Nerd Fonts CSS.

Go★ 120 oct 2026

MIT20 oct 2026 · métricas 2.10.0

npm

34En riesgoíndice de salud