MCP servers and skills: search, filter by where they run, and compare.
Nothing matches every word exactly, so these are the closest matches.
Agent Skill for long-running AI coding tasks: compact task state, context-compaction recovery, session resume, and thread handoff. Works with Codex, Claude Code, WorkBuddy, and DeepSeek Harness. / 面向
SkillMIT★ 1Few starsNo key
Gemba: a task-based system for working with AI agents. The gemba CLI and its specification
SkillMIT★ 3Few starsNo key
Community adaptation of Lauren Tan's pstack for OpenAI Codex: one project-local skill for code investigation, bug fixing, code review and verification.
SkillMIT★ 1Few starsNo key
Nabit Video Recommendations plugin marketplace for Codex. OAuth connection and Git-based updates.
Hosted by authorMITFew starsNo key
Local-first Agent Skill for yesterday's execution traces and verified daily scheduling handoff
SkillMITFew starsNo key
Offline progress dashboards for long-running Claude Code and Codex tasks.
SkillMITFew starsNo key
36 CRM-neutral, evidence-first Agent Skills for safer revenue work.
SkillMITFew starsNo key
多 agent 任务交接协议：一份 CONTEXT.md 交接卡；工程化升三档；交接只发状态不发历史。支持 Qoder / Claude Code / Codex / OpenCode / ZCode。
SkillMITFew starsNo key
One command. The right model for every step. A receipt for the tokens you saved
SkillMIT★ 1Few starsNo key
A Claude skill that picks and combines the right skills for a task without being asked, splits big jobs across agents with their own skills, and learns from what worked.
SkillMIT★ 1Few starsNo key
Label an issue, approve the plan, get a draft PR. A self-hosted GitHub App that runs Claude Code on your repos.
SkillAGPL-3.0★ 6No key
Rules, skills and hooks for running Claude Code, Gemini, Codex and other AI agents on one repo: a manager picks cheap models, delegates and gates risky calls, and a file-based task board prevents coll
SkillMIT★ 1Few starsNo key
Deterministic multi-agent task router & subscription quota scheduler for Claude Code, Codex, Gemini CLI (Antigravity), Grok, and Paseo. Routes sub-agents by model DNA, ability floor, and 5h/7d rate-li
SkillMIT★ 2Few starsNo key
Plans, role-routed subagents, cross-family review, and long-running loops for coding agents in T3 Code.
SkillMIT★ 1Few starsNo key
Agent skill for spec-driven code audits and delegated refactoring: a read-only Lead audits and writes whitelisted specs, Worker subagents implement them in isolated git worktrees. Works with Claude Co
SkillMITFew starsNo key
Task–Architect Interchange — a handoff protocol that automates the transport, never the interchange
SkillMITFew starsNo key
Find, rank, safety-scan and install the best GitHub agent skills (SKILL.md) for any task into Claude Code, Codex, Gemini CLI and OpenCode.
SkillMIT★ 1Few starsNo key
Portable agent skill for exact code retrieval, safe command receipts, and measured context efficiency
SkillMITFew starsNo key
A roadmap in YAML that people and coding agents follow from the command line
SkillMITFew starsNo key
A host-native, runtime-free Skill for AI agents to plan, execute, verify, adapt, and stop with evidence and bounded authority.
SkillMITFew starsNo key
One always-on toolkit for Claude Code and Codex that learns from every session (RSI: recursive self-improvement).
SkillMIT★ 4Few starsNo key
SkillLifeBench: security evaluation across the lifecycle of agent skills. Code, tasks, results and reproducible evidence.
SkillMIT★ 1Few starsNo key
Local Markdown memory your AI coding agents share: Claude Code, Codex, Cursor (preview) and more
PythonMITFew starsNo key
Who stands behind it
