wayfinder
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and processing data from a shared issue tracker, which serves as a major ingestion point for untrusted data that could influence the agent's behavior.
- Ingestion points: The agent retrieves information from issue titles, bodies, and comments on the project's tracker during both the 'Chart the map' and 'Work through the map' phases.
- Boundary markers: There are no instructions in the skill defining strict delimiters or safety wrappers to prevent the agent from interpreting instructions found within ticket descriptions as its own.
- Capability inventory: The skill can create and modify issues, manage Git branches, and launch 'research' subagents, providing an attacker with a vector to execute unintended actions via injected content.
- Sanitization: The skill does not mention any sanitization or verification process for the content it reads from the tracker before using it to drive the logic of its research or planning tasks.
Audit Metadata