skills/spm1001/claude-suite/titans/Gen Agent Trust Hub

titans

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill orchestrates parallel subagents using standard platform features for its intended purpose of code review. No malicious code, exfiltration patterns, or unauthorized system access were detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and analyze untrusted source code.
  • Ingestion points: Scoped source code files and directory context are read into the context of the subagents as defined in the SKILL.md orchestration steps.
  • Boundary markers: The reviewer prompts in references/REVIEWERS.md do not include explicit delimiters or instructions to ignore commands embedded within the code being analyzed.
  • Capability inventory: The skill dispatches high-reasoning subagents using the Task and Explore interface.
  • Sanitization: No sanitization of the input code is performed, as the system relies on the model's analytical persona to maintain its role as a reviewer rather than an executor of the code's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 01:33 AM