project-audit

Stars1

Forks0

UpdatedJuly 22, 2026 at 19:47

Drift analysis — check if deployed skills, steering, and AGENTS.md still match project reality. Use periodically or after major refactors.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Source

smileynet

smileynet/crew-research

View GitHub Repository View Creator Repositories

Download

Related occupationsSOC

Based on SOC occupation classification

Computer Occupations, All OtherComputer and Mathematical Occupations·SOC 15-1299

SKILL.md

readonly

name	project-audit
description	Drift analysis — check if deployed skills, steering, and AGENTS.md still match project reality. Use periodically or after major refactors.
metadata	{"type":"process","invocation":"user-only","practice":null}

Project Audit

Check whether the crew-research deployment still matches project reality.

Checks

1. Verification Commands

Read .kiro/steering/verification-protocol.md
Run each command (build, test, lint) — do they succeed?
Compare against package.json scripts / Makefile / Cargo.toml
If mismatched: update steering file with correct commands

2. AGENTS.md Accuracy

Do listed commands still work? Are any new commands missing?
Do referenced paths (.memory/, .scratch/, .kiro/) exist?
Are listed workflows still available as skills?
Is the project layout description current?

3. Steering Freshness

Do .kiro/steering/ files reference paths/tools that exist?
Is any steering content outdated or contradicted by current practice?

4. Skill Relevance

List deployed skills — are any clearly irrelevant to this project?
Are there recurring tasks where a skill would help but none exists?

5. CONTEXT.md Currency

Are all terms still accurate?
Any terms used in recent work but not yet defined?
Any stale definitions that should be removed?

6. References Directory

Is .references/ used consistently (not resources/ or bare references/)?
If resources/ or references/ exists and is gitignored → flag for rename to .references/
Is .references/ in .gitignore?
Are reference repos documented in AGENTS.md?

7. Unprocessed Decisions

Check for decisions.md, DECISIONS.md, docs/decisions.md
If found: flag for processing into .memory/adr/ (ADR-worthy) or .memory/CONTEXT.md (terms)
Check .memory/decisions.md — should entries be promoted to ADRs?

Output

## Audit Results
- Verification: ✅/❌ (commands work?)
- AGENTS.md: ✅/❌ (accurate?)
- Steering: ✅/❌ (references valid?)
- Skills: N deployed, N relevant, N gaps
- CONTEXT.md: N terms, N stale, N missing
- Config: ✅/❌ (matches reality?)
- References: ✅/❌ (consistent naming, gitignored?)
- Decisions: ✅/❌ (all processed?)

## Actions Taken
- [fixes applied]

## Recommended
- [remaining items needing user decision]

More from this repository

same repository

dispatch-codex-review

smileynet/crew-research

Dispatch Codex to review all unreviewed repository work, then verify its correlated findings ticket or clean result. Use when requesting an independent Codex review, auditing all changes, checking work since the last review, or waiting for Codex findings. Trigger: dispatch codex review, independent review, review all changes, check codex findings, run review agent.

2026-08-011

frontier-work

smileynet/crew-research

Always-on: when a project has tickets, work the frontier (first unblocked open ticket), mark done, update plan status.

2026-08-011

review-new-work

smileynet/crew-research

Incremental and full-history repository review with resumable adoption batches. Use when reviewing new work, initializing review tracking, reviewing all work to date, resuming historical review, inspecting unreviewed commits, or reviewing changed tickets. Trigger: review new work, adopt repository, review all history, resume review, review marker, unreviewed commits.

2026-08-011

handoff

smileynet/crew-research

End-of-session handoff that captures current state for the next session. Use when ending a work session, switching context, or before a long break.

2026-08-011

project-cleanup

smileynet/crew-research

Consolidate project artifacts — promote scratch to memory, deduplicate memory, process decisions, organize scripts, update steering/skills accuracy. Use periodically or when the project feels cluttered.

2026-08-011

eval-harness

smileynet/crew-research

Run and interpret crew-research evals — run.sh flags, scores.jsonl fields, activation TPR/FPR verdicts, resuming interrupted runs. Use when running evals, reading eval results, diagnosing a failing definition, or resuming a dead run. Trigger: run the evals, eval results, scores.jsonl, activation test, TPR, delta threshold, resume the run, skip-completed, known gap.

2026-07-311

Project Audit

Check whether the crew-research deployment still matches project reality.

Checks

1. Verification Commands

Read .kiro/steering/verification-protocol.md

Run each command (build, test, lint) — do they succeed?

Compare against package.json scripts / Makefile / Cargo.toml

If mismatched: update steering file with correct commands

2. AGENTS.md Accuracy

Do listed commands still work? Are any new commands missing?

Do referenced paths (.memory/, .scratch/, .kiro/) exist?

Are listed workflows still available as skills?

Is the project layout description current?

3. Steering Freshness

Do .kiro/steering/ files reference paths/tools that exist?

Is any steering content outdated or contradicted by current practice?

4. Skill Relevance

List deployed skills — are any clearly irrelevant to this project?

Are there recurring tasks where a skill would help but none exists?

5. CONTEXT.md Currency

Are all terms still accurate?

Any terms used in recent work but not yet defined?

Any stale definitions that should be removed?

6. References Directory

Is .references/ used consistently (not resources/ or bare references/)?

If resources/ or references/ exists and is gitignored → flag for rename to .references/

Is .references/ in .gitignore?

Are reference repos documented in AGENTS.md?

7. Unprocessed Decisions

Check for decisions.md, DECISIONS.md, docs/decisions.md

If found: flag for processing into .memory/adr/ (ADR-worthy) or .memory/CONTEXT.md (terms)

Check .memory/decisions.md — should entries be promoted to ADRs?

Output

## Audit Results - Verification: ✅/❌ (commands work?) - AGENTS.md: ✅/❌ (accurate?) - Steering: ✅/❌ (references valid?) - Skills: N deployed, N relevant, N gaps - CONTEXT.md: N terms, N stale, N missing - Config: ✅/❌ (matches reality?) - References: ✅/❌ (consistent naming, gitignored?) - Decisions: ✅/❌ (all processed?) ## Actions Taken - [fixes applied] ## Recommended - [remaining items needing user decision]