analyze
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided markdown files (
spec.md,plan.md,tasks.md,constitution.md) as its primary data sources. These files could potentially contain malicious instructions intended to influence the agent's behavior. However, the risk is mitigated by the skill's strict constraints and limited capabilities. - Ingestion points: Project documents in the feature directory (
spec.md,plan.md,tasks.md) and the project root (constitution.md) are loaded into the agent's context for analysis. - Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are specified for the ingested content.
- Capability inventory: The skill is strictly read-only. It contains no instructions or mechanisms for network access, file writing, subprocess execution, or dynamic code evaluation.
- Sanitization: No input sanitization or filtering is performed on the loaded documentation content prior to semantic modeling.
Audit Metadata