lead

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use git and a task-tracking utility named bd for repository and workflow management.
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection as it processes repository files to provide context for sub-agents.
  • Ingestion points: Reads local project files and directory structures to populate teammate spawn prompts (SKILL.md).
  • Boundary markers: No specific delimiters or safety instructions are prescribed for the passed context.
  • Capability inventory: Can spawn agents, run quality gates, and modify repository state.
  • Sanitization: No content validation or sanitization is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 03:50 PM