audit-docs
Audit a codebase for documentation that has drifted out of sync with the code — stale vocabulary, violated decisions, README/behavior claims the code no…
Skill by krixon
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/krixon-audit-docs
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/audit-docs
What's inside
- Skill name
audit-docs- Description
- Audit a codebase for documentation that has drifted out of sync with the code — stale vocabulary, violated decisions, README/behavior claims the code no longer backs — and surface each as a finding. Static-first, no tool run required. Use when the user wants to check the docs are still accurate, find stale docs, or asks "is the documentation still accurate?" / "what drifted from the docs?".
- License
- MIT
- Instructions
- 41 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Docs Audit
Find where the **prose no longer matches the code** — documented claims the implementation has outgrown — and surface each as a finding. Run the shared audit method in [../AUDIT-METHOD.md](../AUDIT-METHOD.md) with the lens, sub-dimensions, and dimension below.
The aim is *factual drift* — a doc that contradicts the code — not writing quality. A stale claim that misleads a reader or an agent is the target; an awkward sentence that is still true is not.
The project's documentation is both the claims to check *and* the vocabulary finding titles use — whatever the in-context project `CLAUDE.md` points to. With none present, the audit degrades to README and docstring claims; if there is nothing documented to contradict, report nothing rather than invent docs to audit.
## Risk lens: map claims (step 1)
Walk the documentation and collect concrete, checkable assertions — a defined term's meaning, a recorded decision, a README's described command or flag, a docstring's stated behavior. Skip aspirational or roadmap prose: only claims the code is supposed to back right now. Rank by how badly a stale claim would mislead.
## Current state: locate the backing code (step 2)
For each claim, find the code that should make it true. A claim with no corresponding code is itself a signal (dead doc). Trace whether the code *actually* honors the claim — a term still used with the same meaning, an ADR's decision still implemented, a documented flag still wired, the described output still produced. A doc and code merely sitting near each other is not agreement.
## Sub-dimensions to sweep (step 3)
- **Stale vocabulary** — a documented term or definition the code no longer uses, or now uses to mean something else.
- **Violated decision** — a recorded decision (e.g. an ADR) the implementation no longer honors.
- **README / doc behavior** — described behavior, flags, commands, or outputs that no longer match the code.
- **Dead doc** — documentation for a feature, path, or option that no longer exists.
Drop low-confidence noise — a doc that is loose but not wrong is not a finding.
## Dimension (step 4)
- **Dimension** — `docs-drift` (the sub-dimension goes in the title/evidence, e.g. "stale vocabulary term …", "README flag … no longer wired" — never as a new top-level dimension)
- **Suggested category** — usually `enhancement` ("realign the doc with X"); a doc that actively misleads into wrong behavior is a `bug`
- **Where** — the doc and the code it contradicts, named by section / term / module
- **Evidence** — the documented claim, the code that contradicts it, and who it misleads
## Handover
Per [../AUDIT-METHOD.md](../AUDIT-METHOD.md) → *Handover*.
- **artifact:** findings (in [../contracts/finding.md](../contracts/finding.md) shape)
- **default:** `capture` — dedups against open issues, culls, files survivors as `needs-triage`
- **alternatives:** stop (review the findings yourself first)From krixon/skills, MIT.
More from krixon/skills
- 0.502score
auto
krixon
Run a skill workflow unattended — walk the handover chain taking the recommended hop at each step, without asking.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-error-handling
krixon
Sweep a codebase for error-handling defects — swallowed errors, bare catch-alls, and critical calls with no failure path — and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-observability
krixon
Sweep a codebase for critical paths that run blind — no log, metric, or trace at any layer — and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-performance
krixon
Sweep a codebase for performance hazards on hot paths — N+1 data access, per-iteration allocation or IO in loops, unbounded reads, missing caching or…
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-resource-leak
krixon
Sweep a codebase for resource-leak sites — acquired handles, connections, contexts, or locks with no guaranteed release on every path — and surface each as a…
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-coverage
krixon
Audit a codebase for high-risk untested paths and surface them as findings.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-dead-code
krixon
Sweep a codebase for dead code — unreachable branches, never-called functions, orphaned modules — and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-debt
krixon
Harvest in-code debt markers (TODO/FIXME/HACK/XXX and known shortcuts) from a codebase and surface each as a finding, clustered by area.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-security
krixon
Sweep a codebase for security exposure and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-deps
krixon
Sweep a codebase for unhealthy third-party dependencies — outdated majors on load-bearing deps, versions with known advisories, abandoned upstreams, and…
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
deepen
krixon
Find deepening opportunities in a codebase, informed by the project's established domain language and recorded decisions.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
discover
krixon
Grill the user about a product idea before any technical design — challenge that the problem is real, sharpen who it serves, weigh its value against the cost…
SkillCFreshness: C★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 11
- Last commit (this folder)
- Jun 6, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/krixon/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from krixon
- 0.502score
auto
krixon
Run a skill workflow unattended — walk the handover chain taking the recommended hop at each step, without asking.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-observability
krixon
Sweep a codebase for critical paths that run blind — no log, metric, or trace at any layer — and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-performance
krixon
Sweep a codebase for performance hazards on hot paths — N+1 data access, per-iteration allocation or IO in loops, unbounded reads, missing caching or…
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-resource-leak
krixon
Sweep a codebase for resource-leak sites — acquired handles, connections, contexts, or locks with no guaranteed release on every path — and surface each as a…
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-coverage
krixon
Audit a codebase for high-risk untested paths and surface them as findings.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
audit-error-handling
krixon
Sweep a codebase for error-handling defects — swallowed errors, bare catch-alls, and critical calls with no failure path — and surface each as a finding.
SkillCFreshness: C★ 1Claude Code +9
Cite this page
audit-docs skill. Agent Mart. https://agentmart.dev/skills/krixon-audit-docs (last updated 2026-10-24)