repondre

repondre is a command for Claude Code from BenjoNX/Legal. It costs 12 tokens per session (666 once invoked), scanned A, original, MIT.

A command that drafts standard replies to common French legal requests, including privacy-law and contract matters.

In plain words
What is it for?
Use it for requests about data-access rights, document preservation, NDAs, suppliers, French data-protection authority inquiries, formal notices, and data-processing contracts.
Why use it?
It gives teams a starting response while checking whether the situation needs individual legal handling.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

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.

agentmods
npx agentmods add commands/benjonx/legal/repondre
Clone the repo
git clone --depth 1 https://github.com/BenjoNX/Legal

Made for: Claude Code.

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.

agentmods badge for repondre

README.md
[![agentmods](https://agentmods.dev/badge/commands/benjonx/legal/repondre.svg)](https://agentmods.dev/commands/benjonx/legal/repondre)
Your own site
<a href="https://agentmods.dev/commands/benjonx/legal/repondre"><img src="https://agentmods.dev/badge/commands/benjonx/legal/repondre.svg" alt="Measured on agentmods" height="20"></a>
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 666 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00012 $0.00666
Opus 5 $0.00006 $0.00333
Sonnet 5 $0.00002 $0.00133
Haiku 4.5 $0.00001 $0.00067

Measured 5d ago against content hash dd7b89673237, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

repondre 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 5d 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/repondre.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/repondre -- Réponse type juridique

Génère une réponse pour des demandes juridiques courantes en utilisant les modèles configurés. Adapté au droit français (RGPD, Code civil, Code du travail).

Invocation

/repondre [type-demande]

Types de demandes :

  • droit-acces -- Demande d'exercice de droits RGPD (accès, suppression, rectification, portabilité)
  • conservation -- Avis de conservation de documents (litigation hold)
  • nda -- Demande de NDA par une équipe métier
  • fournisseur -- Question juridique d'un fournisseur
  • cnil -- Réponse à un contrôle ou questionnaire CNIL
  • mise-en-demeure -- Réponse à une mise en demeure
  • sous-traitant -- Questions sur le contrat de sous-traitance RGPD

Workflow

Étape 1 : Identifier le type de demande

Si ambigu, afficher les catégories disponibles et demander clarification.

Étape 2 : Vérifier les déclencheurs d'escalade

Avant de générer la réponse, évaluer si la situation nécessite un traitement individualisé :

Déclencheurs universels :

  • Contentieux potentiel ou enquête réglementaire
  • Demande émanant d'une autorité (CNIL, DGCCRF, tribunal)
  • Réponse créant un engagement juridique contraignant
  • Implication d'un dirigeant
  • Médiatisation
  • Situation inédite

Si un déclencheur est détecté : alerter l'utilisateur, recommander l'escalade, marquer tout projet de réponse « POUR REVUE JURIDIQUE UNIQUEMENT ».

Étape 3 : Recueillir les détails

Adapter les questions au type de demande. Pour une demande RGPD par exemple :

  • Nom et coordonnées du demandeur
  • Type de droit exercé (accès, suppression, rectification, portabilité, opposition)
  • Données concernées
  • Délai de réponse (30 jours RGPD, prorogeable de 60 jours)

Étape 4 : Générer la réponse

Utiliser rechercher_legifrance() pour vérifier les textes applicables si nécessaire. Respecter :

  • Le ton approprié (professionnel, clair)
  • Les éléments juridiques requis (références légales, délais)
  • Les prochaines étapes pour le destinataire
  • Les mentions obligatoires (voies de recours CNIL, etc.)

Read the full file on GitHub · 83 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.

  1. 5d ago First seen · 83 lines · 12 tokens per session scan A dd7b89673237

Subscribe to this mod's changes

repondre is a command published in the GitHub repository BenjoNX/Legal (0 stars, last pushed 6mo ago), licensed MIT. It adds 12 tokens to every session and 666 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 2026-08-31.