discover-azure-skills
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- External Data Retrieval: The skill communicates with the GitHub API to fetch plugin and skill metadata from official repositories. This retrieval is restricted to the vendor's own infrastructure, which is a standard practice for discovery-based tools.
- Indirect Prompt Injection Surface: The skill ingests and processes descriptions from external Markdown files from the vendor repository. Ingestion points: Step 3 in SKILL.md fetches content from raw.githubusercontent.com. Boundary markers: No specific delimiters or 'ignore' instructions are provided. Capability inventory: No direct execution, but recommendation of plugin installation. Sanitization: No explicit sanitization of Markdown content.
- Installation Command Guidance: The provided reference files contain instructions for installing plugins, including platform-specific slash commands and the npx package runner. These facilitate extending the environment through official marketplaces.
Audit Metadata