to-roadmap
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from local project files and reflects it into a generated HTML file, creating a surface for potential instruction injection.
- Ingestion points: Reads feature titles, descriptions, and open questions from
./docs/approach.mdand./ideas/CONTEXT.md. - Boundary markers: The instructions lack explicit delimiters or reminders for the agent to ignore embedded instructions in the source files.
- Capability inventory: Capabilities include file system writes (
./ideas/roadmap.html), directory creation (mkdir), and calls to other skills (artifact-scan,skill-feedback-collector). - Sanitization: No explicit sanitization or escaping of the ingested content is specified before it is rendered into the HTML template.
Audit Metadata