code-review-orchestrator

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The orchestrator ingests untrusted repository data and interpolates it directly into subagent prompts, creating a surface for indirect prompt injection. Ingestion points: The skill reads 'diff.patch' and 'code-context.json' (referenced in scripts/launch-subagents.sh) which may contain malicious instructions embedded in code comments or metadata. Boundary markers: The prompts generated in 'scripts/launch-subagents.sh' do not use delimiters or instructions to ignore embedded commands. Capability inventory: The skill launches subagents with the Task tool to perform various analysis tasks. Sanitization: No evidence of input sanitization is present before data is passed to subagents.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:13 PM