wayfinder
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from an external issue tracker to guide its planning and execution logic.
- Ingestion points: The agent is instructed to load map issues and child ticket bodies from a tracker (e.g., GitHub, GitLab) to determine its next steps as seen in the 'Work through the map' section of SKILL.md.
- Boundary markers: The instructions do not define delimiters or specific 'ignore' directives when interpolating content fetched from the issue tracker into the agent's prompt context.
- Capability inventory: The skill allows the agent to execute subprocesses through subagents (/research, /prototype), manage git branches, and modify the issue tracker state.
- Sanitization: There is no evidence of filtering or validation for the data retrieved from the tracker before it is processed by the agent.
Audit Metadata