executing-plans

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. Ingestion points: The skill reads external plan files to obtain instructions for implementation tasks (SKILL.md: Step 1). Boundary markers: The instructions do not specify any delimiters or safety markers to isolate the plan content from the agent's core instructions, relying instead on manual critical review. Capability inventory: The skill grants the agent the ability to execute any task defined within the plan, run verifications, and call other development sub-skills (SKILL.md: Step 2 and 3). Sanitization: There are no explicit instructions for validating, escaping, or filtering the contents of the loaded plan files before the agent follows their instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 12:22 PM
Security Audit — agent-trust-hub — executing-plans