A multi-technology audit command that detects or accepts specified technology stacks and runs a separate review for each one, in parallel or sequence.

Use it to review projects built with stacks such as Symfony or React, set worker and cost limits, and export results to a chosen directory.

It gathers stack-specific findings and can merge them into one report, reducing the need to audit each technology separately.

audit — Command for Claude Code by TheBeardedBearSAS

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo

git clone --depth 1 https://github.com/TheBeardedBearSAS/claude-craft

Made for: Claude Code.

Or install claude-craft, the plugin that ships this one along with the rest of its 56 skills, 94 commands, 47 agents, 5 hooks.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/audit.svg)](https://agentmods.dev/commands/thebeardedbearsas/claude-craft/audit)

Your own site

<a href="https://agentmods.dev/commands/thebeardedbearsas/claude-craft/audit"><img src="https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/audit.svg" alt="Measured on agentmods" height="20"></a>

Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 4,092 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00017	$0.04092
Opus 5	$0.00009	$0.02046
Sonnet 5	$0.00003	$0.00818
Haiku 4.5	$0.00002	$0.00409

Measured 3d ago against content hash 82b80537c42d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

audit 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 3d 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.

.claude/commands/team/audit.md · 377 lines

How it starts

The opening of the file, as written. The whole thing — 377 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Équipe d'Audit Complet - Audit Multi-Technologie Parallèle

Orchestrer un audit complet parallèle sur plusieurs stacks technologiques en utilisant les Claude Code Agent Teams (v2.1.32+). Lance un agent leader (opus) plus N workers stack-auditor (haiku), un par stack technologique détectée, jusqu'à un maximum configurable.

Arguments

$ARGUMENTS

--techs=auto : Auto-détection des technologies (par défaut). Ou spécifier en séparant par des virgules : --techs=symfony,react
--max-workers=4 : Nombre maximum de workers auditeurs en parallèle (par défaut : 4, max : 4)
--output-dir=<path> : Répertoire de sortie personnalisé pour les résultats d'audit
--max-cost=<dollars> : Budget maximum en dollars. Si le coût parallèle estimé dépasse ce seuil, l'exécution est bloquée avec un message OVER BUDGET
--dry-run : Afficher la composition de l'équipe et le coût estimé sans exécuter
--skip-aggregation : Sortir les résultats par stack sans les fusionner
--sequential : Exécuter les audits séquentiellement au lieu de les paralléliser (pas de surcoût Agent Teams). Utile pour les projets mono-technologie ou quand les Agent Teams ne sont pas disponibles.

Prérequis

Claude Code v2.1.32+ avec support Agent Teams
Variable d'environnement CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 définie
Projet avec 2+ stacks technologiques détectées (les projets mono-stack doivent utiliser le flag --sequential)
Tools/AgentTeams/lib/compatibility-check.sh disponible
Tools/AgentTeams/lib/result-aggregator.sh disponible
Tools/AgentTeams/lib/cost-estimator.sh disponible

ℹ️ Ces scripts sont installés automatiquement par claude-craft (make install-agentteams ou via l'installeur). S'ils sont absents, la commande continue en mode dégradé : estimation de coût manuelle et --ralph-mode indisponible (non bloquant).

Garde-Fou Fast Mode (Confirmation Bloquante)

OBLIGATOIRE : Avant de lancer l'équipe, le leader DOIT :

Détecter si le Fast Mode est actif (indicateur lightning bolt dans le terminal)

Si Fast Mode actif :

Afficher le dashboard comparatif standard vs fast via cost-estimator.sh --fast-mode

Afficher un avertissement bloquant avec les coûts comparés :

⚠️  FAST MODE DÉTECTÉ — Coûts Opus 6x plus élevés !

| Mode     | Input ($/M) | Output ($/M) | Coût estimé cet audit |
|----------|-------------|--------------|----------------------|
| Standard | $5.00       | $25.00       | ~$X.XX               |
| Fast     | $30.00      | $150.00      | ~$Y.YY               |

Voulez-vous continuer en Fast Mode ? (oui/non)
Recommandation : tapez /fast pour désactiver avant de continuer.

Attendre la confirmation explicite de l'utilisateur avant de poursuivre
Si l'utilisateur refuse, abandonner avec un message suggérant /fast pour désactiver

Read the full file on GitHub · 377 lines

Changes

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.

3d ago First seen · 377 lines · 17 tokens per session scan A 82b80537c42d

Subscribe to this mod's changes

audit is a command published in the GitHub repository TheBeardedBearSAS/claude-craft (105 stars, last pushed 4d ago), licensed MIT. It adds 17 tokens to every session and 4,092 once invoked, about $0.0001 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 2026-09-03.

Building agents Claude Code Debugging agents ai-tools bmad clean-architecture ddd

audit

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Équipe d'Audit Complet - Audit Multi-Technologie Parallèle

Arguments

Prérequis

Garde-Fou Fast Mode (Confirmation Bloquante)

What this file has done since we first saw it

Other commands, from other repositories

health-check

bootstrap

checklist

clarify

specify

analyze