write-review-artifact

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external draft content from corpus/drafts/{slug}.md. While this ingestion point creates a theoretical surface for indirect prompt injection, the rigid output schema and strict formatting rules provide significant constraints.
  • Ingestion points: Draft files located at corpus/drafts/{slug}.md.
  • Boundary markers: Not explicitly defined in the template.
  • Capability inventory: Writing reviewed artifacts to the ops/technical-reviewer/ directory.
  • Sanitization: No explicit sanitization or escaping of ingested content is mentioned.
  • [SAFE]: No malicious patterns such as remote code execution, obfuscation, or data exfiltration were detected. The instructions are focused on maintaining reporting standards and artifact integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:37 PM
Security Audit — agent-trust-hub — write-review-artifact