planning
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
gh(GitHub) command-line interface to retrieve issue hierarchies and details (SKILL.md). - Evidence: Uses
gh sub-issue list <number>recursively to gather issue metadata. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub issues which could contain malicious instructions.
- Ingestion points: Issue titles, labels, and bodies are read from the GitHub API via the
ghtool (SKILL.md). - Boundary markers: Absent. The instructions do not define delimiters or provide specific guidance to the agent to disregard instructions embedded within issue content.
- Capability inventory: The skill has the ability to write files to the
.docs/plans/directory (SKILL.md). - Sanitization: Absent. Content from issues is analyzed and summarized directly into the generated Markdown plan document without explicit filtering or escaping instructions.
Audit Metadata