Loading...
Loading...
Found 4,171 Skills
Run one autonomous GitHub issue-triage tick and route issues through the canonical states.
Implement one ready-for-agent GitHub issue end to end and open a reviewed pull request.
Measure whether a repository's AI instructions still earn their place, by running the same real task many times with the layer intact and with it stripped, then grading every rule against what actually changed. Runs both arms itself in throwaway git worktrees and never touches the working tree. Agent-agnostic across CLAUDE.md, AGENTS.md, .claude/, .agents/, .cursor/rules, .clinerules, .windsurfrules and copilot-instructions. Use when the user wants to prune, audit, clean up, shrink or "delete" their CLAUDE.md, AGENTS.md, cursor rules, agent instructions or AI layer; when they ask whether their rules are still needed, whether their context is bloated, or what to cut; or when they mention ablating, ablation, or testing their agent without its instructions.
Consolidate Claude Code memory across git worktrees so all branches of the same repo share a single memory directory. Use when memory is isolated per worktree, when switching branches loses context, or when setting up a new machine with worktree-based workflows.
Execute a plan file or Beads epic systematically with git setup, task tracking, quality checks, and commit workflow. Use when implementing a plan, working through a spec, following documented steps, or executing a Beads issue ID (e.g., bd-123, gno-45, app-12).
Generate concise Git commit messages in imperative mood. Analyzes staged changes first; if none, examines unstaged and untracked files. Use when the user asks to create, write, draft, make, or generate a commit message.
Use when entering an unfamiliar codebase, onboarding to a new project, or wanting to assess codebase health before reading code — analyzes git history to reveal hotspots, risk areas, team structure, and development momentum
Control a remote SSH server project from a local git repo with persistent project memory. Use when the user develops locally but runs remotely, wants the agent to understand remote repo mappings across sessions, needs safe local/remote git sync via GitHub, wants to inspect remote state, submit jobs, start interactive sessions, monitor logs, or recover project context at the start of a new coding session.
Fetch review comments from an existing GitHub pull request and implement the requested changes. Use when user says 'address PR feedback', 'implement review comments', 'fix the PR review', 'reviewer asked for changes', or 'my PR got comments'. Do NOT use for initial PR creation (use create-pr) or for writing your own review (use review-diff).
Code commit, PR creation, merge, and issue closure workflow via GitHub CLI (gh). Triggers after a goal (GitHub Issue) implementation is complete — commit code, push branch, create PR, merge, then close the issue. Use when the user says "提交代码", "commit and merge", "创建PR", "合入", "关闭issue", "ship-it", or when a goal implementation is done and code needs to be shipped.
Upgrade flashinfer-python version in TensorRT-LLM. Fetches the latest releases from GitHub (stable and nightly), compares with the current pinned version, lets the user pick a target version, and updates all version references across the repo. Use when the user wants to bump or upgrade flashinfer.
Mid-session save point: create a descriptive git commit and a brief handoff note, then keep working. Use before a risky change or refactor, when switching tasks, or to bank progress without ending the session. Triggers on: /checkpoint, "checkpoint", "save progress", "commit and handoff", "save state", "pause here", "before a risky change". For the full end-of-session ritual with learning extraction, use /wrap-up instead.