dineway-content-brief

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from a prior Research stage, creating a surface where malicious instructions in the input could influence the agent.
  • Ingestion points: The skill reads the Research payload, evidence.json, and findings.md from the Dineway pipeline.
  • Boundary markers: The skill mandates using research as the sole evidence authority and enforces a strict outline schema.
  • Capability inventory: The skill writes a brief.md file and calls a pipeline completion command (content_pipeline_stage_complete).
  • Sanitization: The skill prohibits hallucinations in sensitive categories (medical, legal, financial) and requires authenticated human approval for completion.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 03:11 PM
Security Audit — agent-trust-hub — dineway-content-brief