plan-feature-scaffold
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository files (SPEC.md, roadmap, templates) to generate documentation, which represents an inherent attack surface where instructions embedded in those files could influence agent behavior.
- Ingestion points: Reads feature specifications from
docs/features/, the projectroadmap.md, and SPEC templates. - Boundary markers: Absent; the skill relies on model-based parsing of existing markdown files.
- Capability inventory: Performs file system writes to create artifact sets and updates the roadmap file.
- Sanitization: Not explicitly implemented; the skill fills templates with data derived from discovery steps.
- [EXTERNAL_DOWNLOADS]: The skill allows for a single web research pass when repository documentation is insufficient to answer specific planning questions, which is a controlled information retrieval operation.
Audit Metadata