architect

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by processing untrusted user input (design tasks) and codebase grounding artifacts through a synthesis pipeline involving the arena and how skills. The instructions do not prescribe the use of boundary markers or sanitization, which could allow malicious instructions embedded in the task or codebase to influence the final design.
  • Ingestion points: User-provided architectural tasks and system grounding data processed in SKILL.md.
  • Boundary markers: Absent; no delimiters are specified for untrusted content.
  • Capability inventory: Orchestrates internal tools including how, why, arena, and interrogate.
  • Sanitization: No filtering or validation of external content is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:52 AM
Security Audit — agent-trust-hub — architect