wayfinder
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes data from an external source (the repository issue tracker), including issue bodies and comments. This content is untrusted and could contain malicious instructions designed to influence the agent's behavior during the 'Work through the map' phase.
- Ingestion points: Issue bodies, comments, and label metadata fetched via the issue tracker provider.
- Boundary markers: Absent. The instructions do not specify delimiters or provide warnings to the agent to disregard instructions embedded within the ticket content.
- Capability inventory: The skill can create, update, and close issues, and trigger sub-agents for research or prototyping.
- Sanitization: No evidence of sanitization or content validation before the agent interprets the retrieved ticket data.
- [DATA_EXFILTRATION]: The instructions for 'Task' type tickets suggest recording 'resulting facts' such as 'credentials location' into the resolution comments of issues. If the issue tracker is publicly accessible or shared across a large organization, this guidance could lead to the exposure of sensitive infrastructure metadata or secret locations to unauthorized parties.
Audit Metadata