Claude Skills by WangChangxin0809

Merge an accumulated pile of agent notes, memories, or scratch documents into repository knowledge — freezing the input, synthesizing into a separate output, diffing, and routing each surviving entry to a doc, a guard, a subtree CLAUDE.md, or the small remainder that really is user-scoped. Also covers pruning docs and code the same way. Use this when notes contradict each other, when memory has drifted, when a note references something that no longer exists, when someone asks to clean up or c...

Give a repository a code-and-docs graph an agent can query — tree-sitter symbols, import and call edges, doc-to-code edges, ranked by personalized PageRank from whatever the current task touches. Use this when an agent cannot find the relevant code in a large repo, when grep returns hundreds of hits or none, when someone asks for RAG / semantic search / a code index / an embedding store for their codebase, when onboarding to unfamiliar code, and when you need to know what a change would break.

Build the machinery that enforces a convention instead of documenting it — PreToolUse guards that block an action before it runs, CI gates that judge the worktree, selftests that prove a check can turn red, and structural tests that enforce layering. Use this whenever a rule keeps getting violated despite being written down, whenever adding a lint/check/CI step, whenever asked to enforce architecture boundaries or protect a branch or stop a destructive command, and whenever a check exists but...

Write or restructure documentation that agents and people actually read — choosing which of the four directories it belongs in (how-to, reference, decisions, exec-plans), knowing what belongs in a failure message instead of a file, and keeping the routing table honest. Use this whenever writing a doc, a runbook, a README, an ADR, a design doc, or a plan; whenever someone says the docs are stale, contradictory, ignored, or too long; whenever deciding where a convention or a prohibition should ...

Write the repository's public face — README.md, CONTRIBUTING.md, and the GitHub community health files (SECURITY, CODE_OF_CONDUCT, LICENSE, issue and pull request templates). Use this when creating or rewriting any of those, when deciding whether a sentence belongs in the README or in CONTRIBUTING, when GitHub's Community Standards page reports something missing, when a repository is going public or starting to accept outside contributions, or when someone asks how to make the project underst...

Lay the foundation that makes a repository teach coding agents how to work in it — CLAUDE.md and its subtree files, .claude/ hook wiring, a layered docs/ truth source, scripts/ gates and guards, and a decision record explaining the shape. Use this whenever someone says the agent keeps repeating a mistake, that rules in CLAUDE.md are ignored, that context feels full before they've typed anything, that a new agent or teammate takes too long to get productive, or asks to set up CLAUDE.md / AGENT...