Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Lode Audit

ASecurity

Audits lode/ documentation for content accuracy, checks whether each doc correctly describes the current codebase. Distinct from lode-sync (which only checks timestamps). Produces a staleness table with git-verified findings, then makes targeted updates only on the sections you approve. Use after large refactors, renames, or when the lode feels out of sync with the code. Triggers on: audit lode, lode out of date, lode accuracy check, is the lode correct, lode stale, lode content review, audit...

2 stars
0 votes
0 copies
0 views
Added 10/6/2026
code-qualitygobashtestinggitdocumentation

Security Analysis

A93/100
highPerforms destructive filesystem operations

Pro shows the line behind each finding and how to fix it

Scanned 10/6/2026

$npx -y skills add e128/dotnet-reference --skill lode-audit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Lode Audit?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Lode Audit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/e128-lode-audit/badge)](https://www.skillsdirectory.com/skills/e128-lode-audit)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: lode-audit
description: >
  Audits lode/ documentation for content accuracy, checks whether each doc correctly
  describes the current codebase. Distinct from lode-sync (which only checks timestamps).
  Produces a staleness table with git-verified findings, then makes targeted updates only
  on the sections you approve. Use after large refactors, renames, or when the lode feels
  out of sync with the code.
  Triggers on: audit lode, lode out of date, lode accuracy check,
  is the lode correct, lode stale, lode content review, audit documentation,
  check lode, lode drift, lode wrong, lode inaccurate, lode needs correction.
  Not for: updating a single known lode file (just edit it directly), or checking timestamps
  only (use lode-sync agent).
allowed-tools: Read, Glob, Grep, Bash, Edit, Write, Agent
---

Audit lode/ documentation for content accuracy against the current codebase.

## Phase 0: Resume Check

```bash
cat .claude/tmp/lode-audit/state.md 2>/dev/null
```

If `state.md` exists, load it and skip phases already marked DONE. Otherwise:

```bash
mkdir -p .claude/tmp/lode-audit
```

Initialize `state.md`:
```
# Lode Audit State
Started: {output of scripts/ts.sh}
```

---

## Phase 1: Enumerate Lode Files

Read `lode/lode-map.md` to extract all referenced `.md` file paths. For each path, extract the `*Updated:*` timestamp (format: `YYYY-MM-DDTHH:MM:SSZ`) and record the file path, timestamp (or "no timestamp"), and line count.

Save to `.claude/tmp/lode-audit/files.md`:
```
# Lode Files
| File | Updated | Lines |
|------|---------|-------|
| lode/summary.md | 2026-02-15T00:00:00Z | 42 |
...
```

Write to state.md: `- Phase 1 (Enumerate): DONE — {N} files found -> files.md`

---

## Phase 2: Staleness Assessment

Use `scripts/lode-ts.sh --stale --json` to get a structured staleness report for all lode files in one call. This returns each file's `days_ago` and `commits_since` (commits to src/tests/scripts/.claude since the file's timestamp).

Classification:
- `commits_since > 0` -> **POTENTIALLY STALE**
- No timestamp in file -> **NO TIMESTAMP** (not included in --stale output. Check Phase 1 files.md for these)
- `commits_since == 0` -> **CURRENT**

Save to `.claude/tmp/lode-audit/assessment.md`:
```
# Staleness Assessment
| File | Updated | Commits Since | Status |
|------|---------|---------------|--------|
| lode/dotnet/testing.md | 2026-01-10 | 23 | POTENTIALLY STALE |
...
```

Write to state.md: `- Phase 2 (Assess): DONE — {N} stale, {M} current, {K} no-timestamp -> assessment.md`

---

## Phase 3: Present Findings and Get Approval

Display the assessment table, sorted by: POTENTIALLY STALE first (descending commit count),
then NO TIMESTAMP, then CURRENT.

**Output format:**

```
Lode Audit Results

  # | File                          | Updated    | Commits | Status
----+-------------------------------+------------+---------+------------------
  1 | lode/dotnet/testing.md        | 2026-01-10 |      23 | POTENTIALLY STALE
  2 | lode/infrastructure/ci.md     | 2026-01-28 |      11 | POTENTIALLY STALE
  ...
 12 | lode/summary.md               | (none)     |       — | NO TIMESTAMP
 15 | lode/dotnet/patterns.md       | 2026-03-01 |       0 | CURRENT
```

Ask: **"Which entries do you want to update? Enter numbers (e.g. 1,3,5), 'all stale', or 'none'."**

Wait for response. Save approved list to `.claude/tmp/lode-audit/approved.md`.

Write to state.md: `- Phase 3 (Triage): DONE — user approved {N} files -> approved.md`

---

## Phase 4: Targeted Updates

For each approved file, spawn a targeted agent:

```
Agent: general-purpose, max_turns: 12
Prompt:
  Audit and update the lode file: {file_path}

  Read it, verify its factual claims (class names, method signatures, file paths,
  configuration values, counts, architectural relationships) against current code
  using targeted Grep/Glob/Read — do NOT explore unrelated areas. Report what is
  CORRECT / STALE / MISSING, then apply the minimal edits to make it accurate.
  Preserve existing structure and prose style.
  Update the timestamp: `scripts/lode-ts.sh {file_path}`
```

After each agent completes, record in state.md:
`- Phase 4 ({file_path}): DONE — {summary of changes}`

Report progress as each file completes. After all approved files are done:

---

## Phase 5: Cleanup and Summary

1. Report a final summary table: files updated, key corrections made.
2. Delete checkpoint files:
```bash
rm -rf .claude/tmp/lode-audit/
```
3. Remind: stage the updated lode files before next commit.

---

## Rules

- **Targeted verification only**: each agent reads the relevant code for its file only.
- **Preserve prose style**: corrections should read like the rest of the file.
- **Never auto-approve**: Phase 3 always waits for user input before modifying any file.
- **Commit count is a signal, not a verdict**, POTENTIALLY STALE means "check this", not "definitely wrong".

Attribution

e128e128
View sourceSee grades on GitHubMore from e128 →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman Review

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.

1100021 votes

Caveman Commit

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.

1100021 votes

Springboot Verification

Verification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.

2456590 votes

Verification Loop

一个全面的 Claude Code 会话验证系统。

2456590 votes

Django Verification

Verification loop for Django projects: migrations, linting, tests with coverage, security scans, and deployment readiness checks before release or PR.

2456590 votes
View all in code-quality →