hyperplan
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates unsanitized user planning requests into the prompts for sub-agents, creating a surface for indirect prompt injection.\n
- Ingestion points: User-provided requests are processed in
SKILL.mdand forwarded to team members via orchestration tools.\n - Boundary markers: User requests are wrapped in
<user-request>tags.\n - Capability inventory: Orchestration relies on platform-internal tools including
team_create,team_send_message, and thetasktool. There is no evidence of high-risk tool usage, such as network access or arbitrary shell execution, by the sub-agents or the lead agent.\n - Sanitization: No input validation or escaping is applied to the user-supplied content before interpolation.
Audit Metadata