create-plan
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external requirements to generate implementation plans stored in the repository.
- Ingestion points: The agent ingests 'accepted requirements' and existing source code as context for planning (SKILL.md).
- Boundary markers: The skill does not define explicit delimiters to separate user requirements from internal instructions.
- Capability inventory: The skill performs file-write operations to the 'dev/plans/' directory (SKILL.md).
- Sanitization: The skill contains specific safety instructions to 'Never copy secrets into the plan' and requires 'redaction, disposable data, [and] cleanup' for live verification steps (SKILL.md, references/plan-template.md).
Audit Metadata