From rc
Generates and incrementally refreshes a per-directory codemap (a codemap.md in each significant folder) so agents get a cheap, always-fresh structure map before exploring. Only regenerates maps for directories that changed since the last run, tracked via .rc/codemap state. Use to build or refresh the repo's codemaps; read the maps first in any exploration-heavy task to cut token cost. Do not use to analyze a specific bug or trace one flow (use rc-analyze).
How this skill is triggered — by the user, by Claude, or both
Slash command
/rc:rc-codemapsonnetThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Maintain a hierarchical, per-directory map of the repository so every exploration-heavy task
Maintain a hierarchical, per-directory map of the repository so every exploration-heavy task
starts from a written summary instead of re-scanning source. Each significant directory gets a
codemap.md; a root codemap.md links them. Regeneration is incremental — only directories
that changed since the last run are rebuilt.
codemap.md inside each mapped directory (committed alongside the code)..rc/codemap/last-commit (the git SHA at the last generation).scripts/stale.sh [root] prints the directories that changed since that SHA — or every
directory when there is no state yet (first build). scripts/stale.sh --record [root] records
the current HEAD after a successful generation. Both fail open: on any git problem they
behave as if everything is stale, which is safe.scripts/stale.sh to get the stale directories. If it returns everything,
you're doing a first full build..git, node_modules, vendor, build output, fixtures) — read its files enough to
summarize, then (re)write its codemap.md with: the directory's responsibility, its key
files and their roles, notable patterns, and how it connects to sibling directories. Anchor
claims to real paths.codemap.md so it links every directory map.scripts/stale.sh --record so the next run only rebuilds what changes
after this point.In any task that begins with codebase exploration, read the relevant codemap.md files
first and fall back to grep/Read only for what the maps don't cover. That is the token-saving
payoff — the maps exist so agents don't re-derive structure every session.
stale.sh fails open and reports all directories; do a full build and
skip the --record step (nothing to record without git).npx claudepluginhub rodolfochicone/rc-project --plugin rcNavigate a codebase by map instead of reading files wholesale — a deterministic stdlib script that emits the tree with line counts and top-level symbols, plus the read-the-map-first discipline that cuts exploration tokens by an order of magnitude. Use when asked explore this repo efficiently, stop re-reading the whole codebase, make a map of this project, or which files should the agent actually open. Produces the compact map with its token math (map vs. everything), the navigation discipline, and the open-only-what-matches rule.
Generate a structured, self-maintaining codebase map: a set of atomic Markdown docs in .claude/.codebase-info/ that ground every future Claude session in how the project is built. Use when the user asks to "map the codebase", "document the codebase", "create codebase documentation", "generate architecture docs", "onboard me to this project", "what does this codebase do", "bootstrap codebase docs", "set up codebase-mapper", or "analyze the project structure". Works for any language/stack and for both existing projects and brand-new or empty ones. To refresh an existing map after code changes, use update-codebase-map instead.
Orchestrates parallel subagents to map any codebase, creating docs/CODEBASE_MAP.md with architecture, file roles, dependencies, and navigation. Updates incrementally via git or scans.