traceability-matrix
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests artifact data from external platforms (Jira, GitHub, and local documentation) to build its traceability matrix. This provides a surface for indirect prompt injection where adversarial instructions embedded in requirements or stories could influence the agent's behavior during processing.
- Ingestion points: The skill specifically prompts for artifact inventory sources (Jira, GitHub, documentation folders) in Phase 1.
- Boundary markers: The instructions do not define clear delimiters or "ignore" directives to isolate potentially untrusted external data from the agent's control logic.
- Capability inventory: The skill performs file system write operations (Markdown, CSV) and generates diagrams using external rendering tools.
- Sanitization: There are no specific instructions to sanitize, escape, or validate the content retrieved from external artifact sources.
Audit Metadata