writing-plans
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection by ingesting untrusted user specifications to generate executable plans.
- Ingestion points: User-provided feature specs, requirements, or migration details mentioned in
SKILL.mdandREADME.md. - Boundary markers: Absent. The instructions do not specify delimiters or warnings to ignore instructions embedded within the provided specifications.
- Capability inventory: The skill writes plan documents to
docs/plans/. These plans are structured to contain specific shell commands (e.g.,git commit,pytest) intended for downstream execution by other skills likeexecuting-plansorsubagent-driven-development. - Sanitization: Absent. The skill translates user requirements into a sequence of tasks including code blocks and commands without validating the safety of the resulting plan steps.
- [COMMAND_EXECUTION]: The skill is designed to generate implementation plans that explicitly include shell commands. While the
writing-plansskill itself primarily handles document creation, it generates content specifically designed for automated or semi-automated execution (e.g.,git add,git commit,pytest) in the user's environment.
Audit Metadata