orchestrate

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes a user-supplied goal that is decomposed and distributed to parallel cloud-agent workers.
  • Ingestion points: The parameter in the /ai-eng/orchestrate command within SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore instructions embedded within the user goal are provided.
  • Capability inventory: The skill utilizes the Cursor SDK to spawn cloud agents with repository access and file writing capabilities.
  • Sanitization: There is no evidence of input validation or sanitization of the goal string before it influences worker agent tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:52 AM
Security Audit — agent-trust-hub — orchestrate