Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/TheBeardedBearSAS/claude-craftnpx agentmods add commands/thebeardedbearsas/claude-craft/audit-claude-alignmentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/thebeardedbearsas/claude-craft/audit-claude-alignment)<a href="https://agentmods.dev/commands/thebeardedbearsas/claude-craft/audit-claude-alignment"><img src="https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/audit-claude-alignment/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/thebeardedbearsas/claude-craft/audit-claude-alignment"><img src="https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/audit-claude-alignment.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.03336 |
| Opus 5 | $0.00017 | $0.01668 |
| Sonnet 5 | $0.00007 | $0.00667 |
| Haiku 4.5 | $0.00003 | $0.00334 |
Grade A, and why
audit-claude-alignment scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit d'alignement Claude — claude-craft
Vérifie que claude-craft reste aligné sur l'état réel de l'écosystème Claude : version du CLI Claude Code, modèles et pricing, best practices Anthropic, features de la plateforme, avis de sécurité, et pratiques de la communauté.
Complémentaire de /common:audit-freshness, qui couvre les stacks applicatifs (React, Symfony, Flutter…) et jamais Claude Code lui-même.
Résultat : rapport docs/audit/claude-alignment/<date>.md + <date>.json, et une PR portant uniquement les corrections mécaniques. Tout ce qui demande un jugement part en checklist dans le corps de la PR, jamais en édition automatique.
Arguments
$ARGUMENTS
--quick(défaut) : n'audite que les lentilles dont la source externe a bougé depuis la baseline--full: force les 7 lentilles, efforthighsur sécurité et modèles — cadence mensuelle--lens=<nom>: une seule lentille (cli-release,models-pricing,prompt-context,cc-features,security-cve,community,internal-conformance)--since=YYYY-MM-DD: force la date de baseline — rattrapage après un trou de cadence--dry-run: pré-collecte + liste des agents qui seraient lancés. Aucun agent, aucune écriture--no-pr: rapport seul, ni branche ni PR
MISSION
Étape 0 — Pré-vol et pré-collecte
- Worktree propre obligatoire.
git status --porcelain; si la sortie n'est pas vide, arrêter et le signaler. Des agents d'audit ont déjà pollué le worktree par le passé — on ne construit pas une PR par-dessus des modifications non validées. - Lancer la pré-collecte déterministe (coût LLM nul) :
Elle litnode scripts/collect-claude-signals.mjs [--since=<date>] [--dry-run]config/claude-alignment-baseline.json, interroge le registry npm, les avis de sécurité GitHub, le catalogue communautaire et les pages de doc suivies par empreinte, puis écritdocs/audit/claude-alignment/signals-<date>.json. - Lire ce fichier. Le champ
agents_to_launchest la liste exacte des lentilles à réveiller. Une lentille absente n'est pas auditée : écrire✅ aucun changement depuis <baseline.last_run>dans le rapport, sans lancer d'agent. C'est le principal levier de coût de cette commande. - Calculer l'âge du dernier run (
baseline.last_run→ aujourd'hui). Au-delà de 10 jours, ouvrir le rapport par un avertissement de cadence. - En
--full, ignoreragents_to_launchet retenir les 7 lentilles. En--lens=<nom>, ne retenir que celle-là. En--dry-run, s'arrêter ici en affichant la liste retenue et le coût estimé.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 202 lines · 35 tokens per session scan A 100b565b9386
audit-claude-alignment is a command published in the GitHub repository TheBeardedBearSAS/claude-craft (105 stars, last pushed 6d ago), licensed MIT. It adds 35 tokens to every session and 3,336 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
assemble-team
Assemble a pre-built agent team for parallel work - review, feature, debug, cross-platform, full-stack, or research.
standardize-claude-md
Add missing toolkit sections (Related Global Rules, Quick Start) to existing CLAUDE.md.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.