wayfinder
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a shared issue tracker, creating a surface for potential indirect prompt injection attacks if an attacker can comment on or create issues in the repository.
- Ingestion points: The skill fetches issue titles, bodies, and comments from a repository's issue tracker as specified in SKILL.md.
- Boundary markers: No specific delimiters or 'ignore embedded instructions' warnings are defined for the content retrieved from the tracker.
- Capability inventory: The agent can create, modify, and close issues, and can invoke other skills using the Skill tool.
- Sanitization: The skill does not describe any validation or sanitization logic for the content processed from the external tracker.
- [CREDENTIALS_UNSAFE]: The skill instructions for 'Task' tickets advise the agent to record 'credentials location' in resolution comments on the issue tracker. This practice poses a risk of exposing sensitive infrastructure details or secret storage locations to anyone with access to the issue tracker.
Audit Metadata