Adversarially fact-check DOCS.md files against the actual source code, verifying every concrete claim rather than trusting structure checks. Use when the user asks to audit docs, verify documentation accuracy, check whether docs are still true, or on a scheduled documentation review. This is the semantic layer the mechanical checkers cannot cover.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
scripts/context-drift-check.py proves referential integrity (links, symbols, commands, codemap paths, structure). It cannot prove that prose is TRUE. This skill is the semantic layer: read the doc, read the module, and try to refute every checkable claim.
Given no arguments, audit the DOCS.md files whose reviewed: date is oldest or overdue (the checker's stale warnings list them). Given a module or doc path, audit that doc. A full-repo audit fans out one subagent per doc (they are independent — run them in parallel).
On a scheduled run, guard first — before installing dependencies or reading any docs:
git log --oneline --since='8 days ago' -- crates packages apps '**/docs/DOCS.md' docs/architecture
No output (or only commits whose changes are all outside those paths — CI config, lockfiles, release chores) means nothing meaningful changed: report "no source or docs commits since <date>; audit skipped" and stop. When there are meaningful commits, prioritize the docs whose modules those commits touched.
python3 scripts/context-drift-check.py, node scripts/check-docs-fences.mjs, python3 scripts/check-invariants.py.docs skill, re-run the checkers, bump the doc's reviewed: date (the audit IS the attestation), and commit as docs: ... per the commit skill. Never invent content to fill a finding you did not verify.reviewed: — that is the point of the attestation.No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Canonical rules for writing git commits in the Shift codebase. Use whenever the user asks to commit, stage and commit, create a pull request that requires commits, or draft a commit message. Enforces Conventional Commits, release-note quality, concise subjects, and logical commit boundaries.
The English summary is being prepared. Showing the original description.
Find and remove dead code (unused files, exports, class members) using Knip as a candidate generator, then verify each candidate through AST-level analysis and interface tracing before removing anything. Use when the user asks to clean up unused code, find dead code, or reduce the codebase.
The English summary is being prepared. Showing the original description.
Update or create DOCS.md files for Shift subsystems. Use this skill whenever the user asks to update docs, refresh documentation, create a DOCS.md, write module documentation, or says "update docs for X". Also trigger after completing a large feature when Claude.md says to update docs — check if any DOCS.md in the affected subsystem needs refreshing.
The English summary is being prepared. Showing the original description.
Canonical rules for finding, creating, and updating Shift GitHub issues. Use whenever the user asks to file, create, open, update, triage, or search for an issue, or when substantial work needs an issue before a pull request. Prevents duplicates and defines acceptance criteria and pull-request closure semantics.
The English summary is being prepared. Showing the original description.
Add or revise source-level JSDoc for Shift APIs. Use this skill before writing or editing documentation comments for exported classes, methods, constructors, domain data structures, render frames, reactive state, or any API where caller intent, side effects, lifetime, ownership, or nullability are easy to misunderstand.
The English summary is being prepared. Showing the original description.
How to find and fix performance problems in Shift's desktop app. Use when something is slow, choppy, janky, or laggy (scrubbing, dragging, editing, undo, opening fonts), when profiling or measuring, when adding or reviewing perf tests, and before claiming a change made something faster.
The English summary is being prepared. Showing the original description.