plan-strict
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from GitHub issues and local artifacts to draft implementation plans. It does not provide explicit boundary markers or sanitization for this untrusted data, creating a surface for indirect prompt injection. The impact is mitigated by the skill's specific purpose and limited capability to write only Markdown files in designated plan folders.
- Ingestion points: GitHub issues and project files like
spec.mdandseed.md(specified inSKILL.md). - Boundary markers: No delimiters are mandated for the processing of external context.
- Capability inventory: The skill is restricted to creating folders and writing Markdown files in the
plans/and.pending-decisions/directories. - Sanitization: No sanitization methods are defined for external data.
- [EXTERNAL_DOWNLOADS]: The skill fetches content from GitHub to gather necessary context. This is a legitimate use of a well-known service for project planning.
Audit Metadata