writing-plans
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge between design specifications (untrusted input) and executable implementation plans. This creates a surface for indirect prompt injection where instructions hidden in the design phase could influence the generated output.
- Ingestion points: The skill processes design documents as described in the metadata and instructions in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions within the source design content.
- Capability inventory: The skill facilitates writing files to the local filesystem (docs/plans/) and generates shell commands (such as pytest and git) and code blocks intended for subsequent execution by other skills.
- Sanitization: No validation, escaping, or filtering of the external design content is specified before it is interpolated into the generated plan.
Audit Metadata