component-catalog
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests local instructions, component APIs, tokens, and stories from the codebase. These inputs lack boundary markers or sanitization, creating a surface where malicious instructions in the data could influence the agent's output.\n
- Ingestion points: Step 1 in
SKILL.mdidentifies local instructions, APIs, and stories as data sources.\n - Boundary markers: No markers or 'ignore' instructions are provided to delineate the ingested content.\n
- Capability inventory: The skill generates component specifications, API definitions, and reuse decisions based on analyzed data.\n
- Sanitization: No sanitization or verification logic is present to filter malicious instructions from the inputs.
Audit Metadata