skills/popodidi/harvis/checkpoint/Gen Agent Trust Hub

checkpoint

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses bash for operations like directory creation, file copying, and git-based version control. It also executes a vendor-specific tool 'bd' for session management.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. 1. Ingestion points: project plans (~/.claude/plans/) and external resource metadata. 2. Boundary markers: None identified. 3. Capability inventory: Write, Edit, and Bash access. 4. Sanitization: No explicit sanitization of ingested content before inclusion in metadata. The risk is minimized as data is treated as documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 04:46 PM