Loading...
Loading...
Found 4,877 Skills
Triage GitHub issues by applying type, effort, priority, and area labels. Runs in an isolated context to avoid polluting the main conversation with issue details. Delegates to a specialized triage agent with label validation hooks.
Manage GitHub pull request workflows for coding agents. Use when Codex needs to open, update, monitor, or hand off a PR; wait for CI checks or reviewer feedback; inspect unresolved review threads; address requested changes; summarize PR status; or decide whether to continue, wait, report a timeout, or ask for human input.
This skill should be used when the user wants to run baseline evaluations on existing agent skills, regenerate transcripts after a model upgrade, or check whether a skill still solves the gap it was authored for. Common triggers include "rerun the baselines", "re-eval skill X", "test all the skills", "check for skill drift", and "run the evals". Bakes in verbatim transcript capture (no paraphrasing), deterministic-only grading (regex / contains / file_exists — no LLM-as-judge), and the iteration-N workspace convention. Skip when authoring a new skill (use skill-creator) or modifying skill content directly.
This skill should be used when the user wants to create a new agent skill, scaffold a SKILL.md, validate an existing skill against repo rules, or refactor a skill to match this monorepo's conventions. Common triggers include "build a skill for X", "create a new skill", "scaffold a skill", "add a skill that does Y", "make me a skill", "audit this skill against our rules", and "refactor this skill to match repo conventions". Enforces kebab-case naming, verbatim trigger phrases, selective XML for example boundaries, and a RED→GREEN→REFACTOR cycle. Skip when modifying source code, debugging an existing skill, or writing non-skill markdown.
Pull every meal you ever logged out of MyFitnessPal — per-food CSV, agent-shaped trends, and a local SQLite store. Trigger phrases: `what did I eat this week`, `export my food diary`, `find every time I logged X`, `top foods driving my protein`, `am I hitting my calorie streak`, `use myfitnesspal`, `run myfitnesspal`.
Generate deep research reports on prediction market events using the Octagon Prediction Markets Agent. Combines real-time Kalshi market data with AI-driven analysis to surface price drivers, compare market vs. model probabilities, and identify potential mispricings across 120+ active markets.
Search, install, list, remove, update, or scaffold AI agents with the `agentshq` CLI across many coding CLIs and IDEs. Use when the user wants to discover agents, install them into specific clients, or manage an existing agent catalog.
Extract a validated learning from the current session, store it in the central agent learnings file, and sync the resulting Learnings section into the agent definitions used by the supported CLIs. User-only maintenance workflow for durable agent guidance.
Convert a local AGENT.md into a Claude Code optimized agent. Audits one agent against Claude Code runtime behavior, creates a per-agent DAG rewrite plan with source-backed guardrails, and optionally rewrites the frontmatter and system-prompt body so the agent is thinner, more role-specific, and better aligned with Claude's agent runtime. Use when the user says "convert this agent to Claude", "normalize this AGENT.md", "thin this agent", or "rewrite this persona for Claude Code".
Create implementation task plans in `_/local-plans/<plan-name>.md`. First investigate the codebase using the Explore Agent, then document it in verifiable granularity and parallel-executable units, following the standard format (Background & Purpose, Current Status, Design, File Structure Tree, Implementation Steps, Verification Methods) that can be validated by the plan-verifier Agent. Used for requests like "Make a plan", "Design", "Task decomposition", "Think about implementation approach". plan, planning, design, implementation plan, task decomposition, create-plan
Perform read-only reviews of code changes (`git diff`) for quality, architecture compliance, and security (OWASP Top 10) by delegating to Agent tools. Use for self-reviews before committing/creating PRs, or when requesting "review changes" or "code review". Use implement-review-pr for GitHub PR reviews.
Conventional Commits 1.0.0 + 베스트 프랙티스 워크플로 (diff → staging → type 결정 → secrets blocklist → 사전 체크리스트) + 5 founding principle (atomic / leaves-repo-green / why-over-what / imperative / searchable) + project dialect scaffolding. 커밋을 4 reader (`git log` 스캐너 / `git blame` 추적자 / `git bisect` 사냥꾼 / AI agent — `/clear` 컨텍스트 복원 / PR 리뷰 / changelog 생성 / NL 질의)에게 동시에 도움되는 영구 history로 다룸. 본 파일은 한국어 prose 변형. 룰 자체 (영문 default body, lowercase summary, imperative mood, atomic / why-over-what 등 §0 전 원칙)는 영문 SKILL.md와 동일 — 변형 무관. ALWAYS trigger 조건은 영문 SKILL.md frontmatter §ALWAYS와 동일. Triggers (multi-lingual): EN: commit, git commit, stage, commit message, breaking change, conventional commits, revert, fixup, amend, cherry-pick, changelog KO: 커밋, 깃 커밋, 스테이지, 커밋 메시지, 커밋 룰, 컨벤셔널 커밋, 리버트, 되돌리기, 어맨드, 커밋 컨벤션, 커밋 메시지 검토 JA: コミット, git コミット, ステージ, コミットメッセージ, ブレーキング チェンジ, リバート, アメンド ZH: 提交, git 提交, 暂存, 提交信息, 提交消息, 重大变更, 回滚, 修订 Audience: 한국어를 모국어로 쓰는 개발자. §0 founding principle을 한국어로 먼저 잡고 싶은 사용자에게 적합. §1-§14 룰 자체는 영문 SKILL.md를 정본으로 참조 — 본 변형이 룰을 새로 정의하지 않음.