skill-pour-interroger-judilibre-allison-fiorentino

Fail

Audited by Snyk on Jul 30, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to ask the user for an API key and "coller la clé" into scripts/config.json or pass it via --key, which requires the LLM to accept and embed the secret verbatim in output/files.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). Le workflow requis exécute un client qui appelle l’API Judilibre via scripts/judilibre_client.py (search: envoie le texte de l’utilisateur dans params={"query": query} puis lit le JSON de /search, et decision: lit le JSON de /decision), ce qui implique que du contenu outsider (les termes saisis par l’utilisateur) est ingéré avant traitement par l’agent.

Issues (2)

W007
HIGH

Insecure credential handling detected in skill instructions.

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
HIGH
Analyzed
Jul 30, 2026, 07:35 PM
Issues
2
Security Audit — snyk — skill-pour-interroger-judilibre-allison-fiorentino