wayfinder

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting untrusted content from external issue trackers.
  • Ingestion points: The skill fetches the full body of map issues, child tickets, and resolution comments from an external issue tracker (e.g., GitHub) during the Work through the map phase.
  • Boundary markers: The instructions define a Markdown structure (e.g., ## Destination, ## Question) to separate data, but do not provide explicit instructions to the agent to ignore or isolate potentially malicious instructions embedded within those user-controlled fields.
  • Capability inventory: The skill has the authority to create, update, and close issues on the tracker, and can trigger other capabilities such as /research subagents and /prototype tools.
  • Sanitization: There is no mention of sanitization, filtering, or validation of the data retrieved from the tracker before it is used to influence the agent's next steps or decision-making process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 12:57 PM
Security Audit — agent-trust-hub — wayfinder