agentleFS
cloverinkcloverink/shipwright/skills/audit-full/SKILL.md

audit-full

Skill0 starsChanged 49 days agoMITView on GitHub

Branch-aware audit. On main, scans the whole project and opens one tracker issue. On a feature branch, audits the diff with an Opus auditor, auto-fixes, and stages. Inside /ship it is the audit lens of the parallel A/R batch.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. HAX-Studio/mnemonica★ 3same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository12 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No commit: whoever called this commits per phase.

    line 84

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/audit-full

Audit that changes behavior based on where you are. See Branch-Aware Mode.

Modes

WhereModeBehavior
mainStandaloneScan the whole project → open ONE consolidated GitHub issue
feature branchIn-ticketAudit the diff vs main → auto-fix → stage (no commit)
inside /shipLens/ship spawns the auditor next to the reviewers; this file is only the checklist

Execution

The audit itself is judged by the read-only code-reviewer agent (Opus) with lens audit, named auditor, writing to a findings file. This session applies the fixes. The auditor never edits.

Large project on main? Split by top-level area and spawn one auditor per area in one tool-call block.

Spawn (standalone modes). Run the setup as one Bash call and paste the echoed values as literals, since shell variables do not survive between Bash calls:

RUN_DIR="${TMPDIR:-/tmp}/shipwright/$(git branch --show-current | tr / -)-$(date +%s)"
mkdir -p "$RUN_DIR" && printf 'STATUS: RUNNING\n' > "$RUN_DIR/audit.md" && echo "RUN_DIR=$RUN_DIR"
Agent({
  subagent_type: "code-reviewer",          // "shipwright:code-reviewer" when installed as a plugin
  name: "auditor",
  prompt: "Lens: audit. Checklist: <SKILLS_DIR>/audit-full/SKILL.md §Audit checks.
           Scope: <whole project | changed files vs BASE>. Findings file: <RUN_DIR>/audit.md.
           Append, flip line 1 to STATUS: DONE, reply with path + verdict only."
})

SKILLS_DIR is the parent of this skill's base directory. The agent has no Skill tool, so the checklist path must be absolute. One auditor per area → one findings file per area (audit-<area>.md).

Audit checks

The audit lens checklist. The auditor loads this section.

  1. Config (only if config files changed): scripts that point nowhere, duplicate or conflicting settings, secrets in config, CI steps that no longer match the code
  2. Code: dead code, unused exports, magic values, duplicated logic, functions over ~50 lines, nesting over 3, accidental O(n²) on unbounded input
  3. Tests: slow tests, inline mocks that should be shared fixtures, missing edge cases (empty, error, boundary), tests that assert nothing
<!-- CONFIGURE: add your own audit checks or remove passes you don't need -->

Priorities

PriorityMeaningAuto-fixed in-ticket?
P0 Criticalsecurity, data lossyes
P1 Majorbug, perf regressionyes
P2 Warningconvention violationyes
P3 Suggestionstyle, naming nitno, listed in the report (inside /ship: in the PR body)

Standalone (on main)

  1. Spawn auditor(s) over the whole project
  2. Group findings by module
  3. Open ONE GitHub issue: Audit: <date> · <N> findings across <M> areas

One issue, not twenty: one consolidated tracker is easier to triage and does not spam the repo.

In-ticket (feature branch)

  1. BASE=$(git merge-base origin/main HEAD) → changed files only
  2. Spawn the auditor over them
  3. Fix P0-P2 → lint → git add (no commit)
  4. Report fixed vs needs-manual-attention

No commit: whoever called this commits per phase. See Phase Bundling.

Usage

/audit-full              # mode detected from the current branch

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.