recoup-internal-consulting-skill-packager
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified. The skill is designed for internal administrative tasks and includes proactive security measures such as credential sanitization and environment path removal.
- [INDIRECT_PROMPT_INJECTION]: The skill processes files from repositories, creating an attack surface for indirect prompt injection. The skill mitigates this through mandatory review and sanitization instructions. 1. Ingestion points: Plugin source repository files, manifests, and Git status (SKILL.md). 2. Boundary markers: Step 4 requires manual review of a snapshot for a new sharing boundary. 3. Capability inventory: The skill allows file reads, manifest updates, Git commits, and publication to remote repositories. 4. Sanitization: Instructions explicitly mandate checking for credentials, local paths, and private data.
Audit Metadata