wayfinder
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core loop of reading and updating shared issue tracker data.\n
- Ingestion points: The agent fetches and reads the map issue and child ticket bodies from a repository issue tracker to determine planning steps and decisions (SKILL.md).\n
- Boundary markers: The instructions lack explicit delimiters or warnings to the agent to ignore or isolate potentially malicious instructions embedded in the issue tracker content.\n
- Capability inventory: The agent has the ability to create new issues, post resolution comments, close issues, and modify the central "map" issue body, which could be exploited if malicious instructions are processed (SKILL.md).\n
- Sanitization: There is no evidence of sanitization or validation of the content retrieved from the tracker before it is used to influence the agent's planning logic.
Audit Metadata