datagouv-apis

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation for official data.gouv.fr APIs. All endpoints and examples are consistent with the stated purpose of open data access.
  • [EXTERNAL_DOWNLOADS]: References to external documentation (guides.data.gouv.fr) and development tools (github.com/datagouv, github.com/etalab) point to official and author-verified sources.
  • [CREDENTIALS_UNSAFE]: The skill includes explicit security warnings to avoid logging or exposing sensitive API keys during write operations.
  • [PROMPT_INJECTION]: The skill processes untrusted user-generated content from the open data catalog, which is a potential surface for indirect prompt injection.
  • Ingestion points: Catalog search (/datasets/) and tabular data retrieval (/resources/{rid}/data/).
  • Boundary markers: Not specified.
  • Capability inventory: No dangerous tool access or execution capabilities are requested by the skill instructions.
  • Sanitization: Not addressed in the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 05:15 PM
Security Audit — agent-trust-hub — datagouv-apis