skill-search
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill performs legitimate network operations to fetch metadata from its associated service.
- Evidence: The script scripts/search_catalog.py uses the requests library to perform HTTP GET requests to the skillscatalog.ai API endpoints for searching and retrieving skill details.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the catalog API, which is a potential surface for indirect injection if catalog content is malicious.
- Ingestion points: API responses containing skill names, descriptions, and vendor information are processed in scripts/search_catalog.py.
- Boundary markers: The skill presents retrieved data directly to the agent output without specific delimiters or isolation warnings.
- Capability inventory: The skill is restricted to fetching and printing text; it contains no logic for file system modification, command execution, or further network exfiltration based on ingested content.
- Sanitization: The script implements basic string truncation for skill descriptions to limit output length.
Audit Metadata