Skills
Navigate
skills/prowler-cloud/prowler/prowler-attack-paths-query/Gen Agent Trust Hub

prowler-attack-paths-query

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches attack path reference data from DataDog's official GitHub repository and graph schema documentation from Prowler Cloud's GitHub repository.
  • Evidence: curl -s https://raw.githubusercontent.com/DataDog/pathfinding.cloud/main/docs/paths.json in SKILL.md.
  • Evidence: References to schema files at https://raw.githubusercontent.com/prowler-cloud/cartography/ in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides examples for fetching and processing JSON data from external public repositories. This establishes an ingestion surface for external content that could theoretically contain instructions intended to influence the agent's behavior.
  • Ingestion points: SKILL.md (Input sources section).
  • Boundary markers: None.
  • Capability inventory: The skill utilizes Bash and WebFetch tools to retrieve and process data.
  • Sanitization: Data is filtered via jq to select specific fields, providing structural validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 08:34 PM