wayfinder
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes instructions and data from external GitHub issues (titles, bodies, and comments) which could be controlled by third parties. This creates an attack surface for indirect prompt injection.\n
- Ingestion points: GitHub issues (metadata and comments) via the
ghtool as described in SKILL.md.\n - Boundary markers: The instructions do not define delimiters for separating issue content from agent logic.\n
- Capability inventory: The skill can create issues, perform Git commits and pushes, open pull requests, and invoke subagents.\n
- Sanitization: There is no mention of input sanitization or validation for issue-derived content.\n- [COMMAND_EXECUTION]: The agent uses the
ghCLI andgitto manage project state and documentation. These commands are necessary for the skill's stated goal of issue-based project navigation.\n- [SAFE]: The skill contains no obfuscation, data exfiltration patterns, or unauthorized persistence mechanisms. It follows standard automation practices for GitHub-based workflows.
Audit Metadata