code-flow-tracer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes code and data flows which could contain untrusted content. Instructions hidden within the code being traced (e.g., in comments) might attempt to subvert the agent's logic during the analysis process.
- Ingestion points: Processes codebase content through the
/sourceatlas:flowtool based on user queries. - Boundary markers: No explicit delimiters or warnings to ignore instructions within the analyzed data are provided.
- Capability inventory: Invokes the
sourceatlas:flowtool to generate call graphs and flow visualizations. - Sanitization: No sanitization or validation of the analyzed code content is described in the skill instructions.
Audit Metadata