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.
npx skills add mrstev3n/balise-skills --skill balise-web-legalgit clone --depth 1 https://github.com/mrstev3n/balise-skillsWrote 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/skills/mrstev3n/balise-skills/balise-web-legal)<a href="https://agentmods.dev/skills/mrstev3n/balise-skills/balise-web-legal"><img src="https://agentmods.dev/badge/skills/mrstev3n/balise-skills/balise-web-legal/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/skills/mrstev3n/balise-skills/balise-web-legal"><img src="https://agentmods.dev/badge/skills/mrstev3n/balise-skills/balise-web-legal.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 | $0.00171 | $0.02043 |
| Opus 5 | $0.00086 | $0.01022 |
| Sonnet 5 | $0.00034 | $0.00409 |
| Haiku 4.5 | $0.00017 | $0.00204 |
Grade A, and why
balise-web-legal 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 12d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conformité juridique des sites
Travailler comme assistant de recherche, d'audit et de rédaction. Partir du produit et des traitements réels, puis déterminer les juridictions et documents requis. Une page juridique ne prouve jamais, à elle seule, la conformité opérationnelle.
Verrou préalable obligatoire
Pour chaque nouveau site, produit, service ou audit, interrompre le traitement avant toute inspection, recherche, analyse de conformité ou rédaction substantielle. Dans la première réponse, poser le questionnaire de cadrage ci-dessous et ne fournir ni audit, ni conclusion de conformité, ni document juridique dans ce même message.
Appliquer ce verrou même si les réponses semblent déjà présentes dans la conversation, les pièces, le code ou le site fourni. Ne pas préremplir, déduire ni confirmer ces réponses à la place de l'utilisateur. L'utilisateur doit répondre lui-même ou autoriser explicitement l'agent à utiliser les informations déjà connues. Une formule comme « utilise ce que tu sais déjà » permet alors de construire la fiche depuis le contexte, de signaler les informations incertaines et de ne demander que les éléments encore manquants.
Poser les questions en une seule série courte :
- Quel résultat attendez-vous : diagnostic initial, audit documenté, rédaction, correction d'interface, plan d'implémentation ou préparation à une validation juridique ?
- Quelle entité représentez-vous et quel est son rôle dans le service ?
- Quelles entités éditent, exploitent, vendent ou fournissent le service ? Pour chacune, indiquez l'identité, la forme ou le statut, le pays d'établissement et le rôle.
- Quels pays sont réellement ciblés et quelles catégories de personnes utilisent le service ?
- Quel est l'état du produit : conception, test, accès restreint ou service public déjà exploité ? Quelles fonctions impliquent comptes, formulaires, paiements, communications, données personnelles ou traceurs ?
- Quels prestataires, hébergeurs, outils de mesure, moyens de paiement ou transferts internationaux sont connus ?
- Quels documents, registres, contrats, configurations ou preuves sont disponibles ? Existe-t-il une date de lancement, de publication, de contrôle ou une autre échéance ?
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/applicability-router.md 2.2 KB
- references/audit-and-evidence.md 2.1 KB
- references/cookies-and-interface.md 2.0 KB
- references/document-set.md 1.9 KB
- references/international-overlays.md 1.7 KB
- references/jurisdictions-central-africa.md 2.7 KB
- references/jurisdictions-comoros.md 1.1 KB
- references/jurisdictions-eu-france.md 4.3 KB
- references/jurisdictions-west-africa.md 3.6 KB
- references/legal-notices-ecommerce.md 1.9 KB
- references/privacy-notice.md 2.4 KB
- references/verification-and-delivery.md 2.3 KB
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.
- 12d ago First seen · 86 lines · 171 tokens per session scan A 4eda5ebf9280
balise-web-legal is a skill published in the GitHub repository mrstev3n/balise-skills (11 stars, last pushed 23d ago), licensed Apache-2.0. It adds 171 tokens to every session and 2,043 once invoked, about $0.0009 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 skills, from other repositories
compliance
An entry check and compliance-validation skill for a process with PC0–PC10 checks and FC, SC, RC, and T rules. The description does not explain what those abbreviations mean.
privacy-compliance-architecture
A guide for building systems that handle personal information and privacy obligations. It covers consent, data minimisation, retention and deletion, data location, anonymised logs, and requests to access, export, or delete data.
banking-workflow-assistant
Use when bank operations, relationship managers, or credit middle-office teams need to pre-check onboarding, credit, KYB, due diligence, or post-loan documents, route exceptions, track SLA risk, and preserve audit-ready compliance reasoning.
pr-compliance-reviewer
Use when the parent pipeline routes a compliance-related PR review request. Reads the diff and emits structured complianceFindings using the compliance specialist heuristics.
pr-licensing-reviewer
Use when the parent pipeline routes a licensing-related PR review request. Reads the diff and emits structured licensingFindings using the licensing specialist heuristics.
pr-gdpr-reviewer
Use when the parent pipeline routes a gdpr-related PR review request. Reads the diff and emits structured gdprFindings using the gdpr specialist heuristics.