how-to-delegate

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it processes user input and passes it to sub-agents during delegation.
  • Ingestion points: The skill ingests user tasks via the $ARGUMENTS placeholder within <user_instructions> tags in SKILL.md.
  • Boundary markers: User input is encapsulated within XML-style tags.
  • Capability inventory: The skill utilizes the Agent() tool in Step 10 to execute delegated tasks and the Skill() tool in Step 1 to load environment dependencies.
  • Sanitization: The skill implements a 'Pre-Flight Verification' checklist (Step 8) that requires the agent to audit the generated prompt for factual accuracy, remove assumptions, and ensure implementation details are not prematurely prescribed before delegation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:40 AM