performance-expert
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes performance architecture files and APM metrics from the workspace to provide diagnostic recommendations. This creates a surface where malicious instructions embedded in these files (e.g., inside a performance report) could influence the agent's output.
- Ingestion points: Project context discovery reads architecture and performance files from .agents/memory/ and the workspace.
- Boundary markers: The skill instructions do not define delimiters or warnings to ignore instructions embedded in the analyzed data.
- Capability inventory: The skill prescribes CLI commands (k6, node) and suggests code modifications to the caller.
- Sanitization: No explicit validation or filtering of external configuration content is defined.
Audit Metadata