wayfinder

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to manage and execute work based on content retrieved from an external issue tracker (e.g., GitHub Issues).
  • Ingestion points: The agent fetches issue bodies, titles, and comments via tracker-specific queries to build its 'Map' and 'Tickets'.
  • Boundary markers: The instructions lack explicit requirements for the agent to use delimiters or ignore instructions embedded within the ticket content when processing or summarizing them.
  • Capability inventory: The skill has the capability to create/edit issues, close tickets, and invoke other functional skills such as /research, /prototype, and /domain-modeling to perform actions.
  • Sanitization: There is no mention of sanitizing or validating the content of the issues before the agent acts upon the 'Questions' or 'Tasks' defined in them. An attacker with access to the issue tracker could potentially inject malicious instructions into a ticket that the agent would then execute autonomously.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 02:31 PM