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 Ludovic33Fr/product-ai-toolbox --skill feature-ethics-auditgit clone --depth 1 https://github.com/Ludovic33Fr/product-ai-toolboxWrote 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/ludovic33fr/product-ai-toolbox/feature-ethics-audit)<a href="https://agentmods.dev/skills/ludovic33fr/product-ai-toolbox/feature-ethics-audit"><img src="https://agentmods.dev/badge/skills/ludovic33fr/product-ai-toolbox/feature-ethics-audit/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/ludovic33fr/product-ai-toolbox/feature-ethics-audit"><img src="https://agentmods.dev/badge/skills/ludovic33fr/product-ai-toolbox/feature-ethics-audit.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 | $0.00058 | $0.01665 |
| Opus 5 | $0.00029 | $0.00833 |
| Sonnet 5 | $0.00012 | $0.00333 |
| Haiku 4.5 | $0.00006 | $0.00167 |
Grade A, and why
feature-ethics-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 11d 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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
feature-ethics-audit
Fiche éditoriale
Objectif. Auditer une feature envisagée sur ses risques éthiques (manipulation, biais, données).
Entrées. Description fonctionnelle, profil utilisateur cible, mécanisme de la feature.
Sorties. Diagnostic d'exploitation de vulnérabilité, alternatives moins exposées, recommandation explicite.
Cadence d'usage. À toute proposition de feature à fort potentiel d'engagement.
Mode opératoire
Quand m'invoquer
L'utilisateur me soumet une feature à auditer, particulièrement si elle vise l'engagement, la rétention, la conversion, ou si elle traite des données utilisateurs sensibles. Il me fournit la description fonctionnelle, le profil cible, et le mécanisme (ce qui s'active, quand, comment).
Procédure
- Reformuler la feature sans interpréter, en isolant le mécanisme central.
- Identifier la vulnérabilité activée : quel biais cognitif, quel besoin émotionnel, quelle asymétrie d'information la feature mobilise pour fonctionner ? Toute feature à fort engagement en active au moins un.
- Évaluer l'intensité : la feature est-elle conçue pour rendre l'arrêt difficile (sunk cost, FOMO, intermittence variable), ou laisse-t-elle l'utilisateur libre de partir sans coût psychologique ?
- Identifier les segments à risque : quels utilisateurs sont disproportionnellement exposés (mineurs, personnes vulnérables, utilisateurs en situation de stress, biais socio-économiques) ?
- Vérifier les données : la feature collecte/utilise-t-elle des données nouvelles ? Sont-elles strictement nécessaires ? Sont-elles cohérentes avec la finalité initialement consentie ?
- Générer 2-3 alternatives qui atteignent ~80% du bénéfice business sans le mécanisme problématique.
- Recommandation explicite : continuer / continuer avec garde-fous / ne pas continuer. Justifier en 2-3 phrases.
Format de sortie
# Audit éthique — {nom de la feature}
## Description reformulée
> {description en 2-3 phrases, neutre}
## Vulnérabilité activée
- **Biais ou besoin mobilisé** : {nom du biais ou besoin}
- **Intensité d'activation** : faible / moyenne / forte
- **Mécanisme** : {comment la feature exploite ce biais}
## Segments à risque
- {segment 1} — {raison de l'exposition disproportionnée}
- {segment 2} — ...
## Données
- Données collectées : {liste}
- Données strictement nécessaires : {liste}
- Cohérence avec la finalité consentie : oui / partielle / non
## Drapeaux levés
- {drapeau 1}
- {drapeau 2}
## Alternatives moins exposées
### A1 — {nom de l'alternative}
{description courte}
Bénéfice business attendu vs original : ~{x}%
Risque éthique vs original : {réduit / similaire / aggravé}
### A2 — ...
## Recommandation
**{continuer | continuer avec garde-fous | ne pas continuer}.**
{2-3 phrases qui justifient. Si "continuer avec garde-fous", lister les garde-fous obligatoires.}
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.
- 11d ago First seen · 144 lines · 58 tokens per session scan A aa15727f73ee
feature-ethics-audit is a skill published in the GitHub repository Ludovic33Fr/product-ai-toolbox (1 stars, last pushed 4mo ago), licensed MIT. It adds 58 tokens to every session and 1,665 once invoked, about $0.0003 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.
Other skills, from other repositories
journey-simulation
Use when caller wants to observe how a stranger encounters a flow, artifact, or sandbox — triggers like "simulate a user journey", "test our onboarding / checkout / signup", "will my ICP convert", "how does a cold reader experience this README", "first-time user test", "cognitive walkthrough", or any request to…
check-docs-consistency
Cross-reference project docs for drift, stale references, and contradictions. Outputs timestamped report. Discovers Markdown wherever it lives in the repo.
memory-graveyard
Audit which entries in your project's MEMORY.md Claude actually reads. Use when the user wants to find dead memory entries, broken pointers in MEMORY.md, orphan memory files, or surface entries that fall below the system-prompt truncation cutoff. Triggers on "audit MEMORY.md", "dead memory entries", "broken memory…
skill-graveyard
Audit which Claude Code skills you actually invoke. Use when the user wants to find dead/unused skills, hallucinated skill invocations, per-project skill stats, or token cost of installed skill metadata. Triggers on "what skills don't I use?", "clean up my skills", "why didn't this skill work?", "audit skills". Runs…
mcp-graveyard
Audit which MCP server tools your Claude Code sessions actually invoke. Use when the user wants to find dead MCP servers (configured but never called), hallucinated MCP tool calls, or per-project MCP usage. Triggers on "which MCP servers don't I use?", "clean up MCP config", "remove unused MCP servers", "audit MCP…
semble-setup
Project setup tools for semblecode semantic code search. Semantic search finds code by meaning or concept, rather than only matching exact words.