Skip to content
Skills Library
Programming · Writing

codemap

/codemap

Use for evidence-based codebase documentation. Walks the repo, builds a map of components + responsibilities + interconnections. Output is hand-curated markdown with diagrams. Variants — codemap-fast, codemap-diff, codemap-trust, codemap-export.

Repository
sift-coder ↗
Licence
MIT
Last repository change
03.06.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 03.06.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Use for evidence-based codebase documentation. Walks the repo, builds a map of components + responsibilities + interconnections. Output is hand-curated markdown with diagrams. Variants — codemap-fast, codemap-diff, codemap-trust, codemap-export.

Original author description

Use for evidence-based codebase documentation. Walks the repo, builds a map of components + responsibilities + interconnections. Output is hand-curated markdown with diagrams. Variants — codemap-fast, codemap-diff, codemap-trust, codemap-export.

How it works

  1. Variants
  2. Method (full codemap)
  3. Output shape
  4. Rules
  5. Anti-patterns
  6. When NOT to use
  7. Subagent dispatch
  8. Value over native CC
Original SKILL.md headings · 108 lines, 464 words
SKILL.md preview
---
name: codemap
description: Use for evidence-based codebase documentation. Walks the repo, builds a map of components + responsibilities + interconnections. Output is hand-curated markdown with diagrams. Variants — codemap-fast, codemap-diff, codemap-trust, codemap-export.
---

# codemap

Evidence-based codebase doc. Map of what exists, who depends on what, what each module does. Distinguishes confidence levels (proven by tests / inferred from usage / guessed).

## Variants

- **codemap** (this skill) — full pass, single repo
- **codemap-fast** — quick structural scan, no deep semantics
- **codemap-diff** — what changed between two refs (semantically, not just diff)
- **codemap-trust** — confidence/risk per module (which areas to trust, which to verify)
- **codemap-export** — machine-readable JSON output for tools

## Method (full codemap)

1. **Top-down survey.** Top-level dirs; one-line responsibility for each.
2. **Module inventory.** For each subdir:
   - LOC
   - Test coverage
   - Last-touched dates
   - Public API (exports / endpoints)
   - Internal deps (what does it import?)
   - External deps (npm packages used)
3. **Dependency graph.** Module-level. Mermaid.
4. **Hotspots.** High-churn × low-coverage = risk.
5. **Confidence labels** per module — high (tests + recent), med (works but uncovered), low (works but unstable).
6. **Output: ARCHITECTURE.md or codemap.md.**

## Output shape

```
# Codemap: <repo name>

## Top-level
- src/        — code
- tests/      — tests
- docs/       — hand-curated docs
- ...

## Modules

### src/memory/  (5,300 LOC, 96.86% cov, last-touched 2026-04-30)
**Responsibility:** persistent memory engine — daemon, MCP server, retrieval

**Public API:**
  - daemon: UDS server for capture
  - mcp/server: stdio MCP server
  - storage: dual-backend SQLite

**Imports from:** src/core, src/utils
**Imported by:** src/services (none yet)
**External deps:** better-sqlite3, node-sqlite3-wasm, gpt-tokenizer

**Confidence:** high (518 tests, recent)

### src/core/  (200 LOC, 100% cov, last-touched 2026-05-01)
**Responsibility:** plugin-wide primitives — config, paths, errors, logger
...

## Dependency graph

\`\`\`mermaid
graph TD
  Memory --> Core
  Memory --> Utils
  Services --> Core
  Services --> Utils
\`\`\`

## Hotspots
- (none — coverage strong, churn matched by tests)

## Confidence summary
- High:  4 modules (memory, core, utils, services)
- Med:   1 module (formatters — light usage)
- Low:   0 modules
```

## Rules

- **Cite LOC + coverage from real measurement,** not estimate.
- **Confidence labels with reasoning.** "High because <evidence>".
- **Hotspots are evidence-cited** — high churn from `git log`, low coverage from coverage report.
- **Don't bloat with file lists.** Module-level, not file-level.

## Anti-patterns

- File-by-file listing (use a tree command for that)
- Hotspots without evidence (just guessing)
- Confidence labels without reasoning
- Out-of-date diagrams (use `/codemap-diff` to verify)

## When NOT to use

- Tiny repo (manual scan faster)
- One module (just read it)
- Auto-generated doc tool already in place — augment, don't replace

## Subagent dispatch

- `Explore` for the survey
- `Bash` for `git log --shortstat`, `wc -l`, coverage extraction
- `general-purpose` for the synthesis

## Value over native CC

CC will list files. CC won't naturally produce evidence-cited confidence labels, dependency graphs, hotspot analysis. The evidence-based shape IS the value.

Versions

RepositoryScoreLicence
This page · ialameh/sift-coder100MIT
jordancoin/codemap100MIT
AZidan/archflow100MIT
alvinunreal/oh-my-opencode-slim100MIT
jordancoin/codemap100MIT
ils15/copilot-global-config100MIT
azidan/codemap100MIT
jellydn/my-ai-tools96MIT
QDinvest/Code-Map-QD96MIT

Database snapshot: 13.11.2026 · package version cfb554cc360e (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100codemap-export
    Use to export the codemap as machine-readable JSON for downstream tooling. Same data as /codemap, different format. "Export the codemap", "give me the codemap as JSON".
  • 90rc-codemap
    Builds and refreshes per-directory codemap.md as a cheap structure map. Use before exploring a large repo. Do not use to trace one bug.
  • 100codemap-trust
    Use to assess confidence/risk per module. "Which parts of this codebase can I trust", "where's the risk concentrated". Outputs a trust map with reasoning.
  • 100codemap-refresh
    Use when code has changed since codemap files were generated and the user wants them brought up to date: "refresh", "update the diagrams", "docs are stale". Detects which docs/codemap files are outdated and regenerates only those, preserving manual edits.
  • 100config-setup
    Set up or tune .codemap/config.json so Codemap focuses on code-relevant parts of the repo. Use when config is missing, boilerplate, noisy, or mismatched to the stack.
  • 90explore
    Systematic codebase exploration using codemap's analysis tools. Use when understanding how code works, tracing dependencies, or onboarding to a new area.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.