wayfinder
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process content from external files and issue tracking systems, which creates an attack surface for instructions embedded in that data.
- Ingestion points: The workflow retrieves information from local files like CLAUDE.md or AGENTS.md and performs queries against external issue trackers to read decision tickets.
- Boundary markers: The instructions do not include explicit delimiters or guardrails to ensure the agent disregards instructions that might be contained within the retrieved ticket text.
- Capability inventory: The agent uses tools such as /research, /prototype, /domain-modeling, and /to-spec, which execute based on the context provided by these ingested sources.
- Sanitization: There is no requirement or logic specified for the validation or sanitization of input retrieved from external files or trackers.
Audit Metadata