dineway-content-writer

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted research data and briefs to generate article content, creating a surface for indirect prompt injection.\n
  • Ingestion points: Reads artifact contracts and Research/Brief results from the local filesystem at ../dineway-content-pipeline/ (SKILL.md).\n
  • Boundary markers: Implements structural validation and requires factual claims to map to Site Context evidence (SKILL.md).\n
  • Capability inventory: Executes dineway content CLI commands and writes to project-local directories at .dineway/ (SKILL.md).\n
  • Sanitization: Explicitly forbids the inclusion of prompts or internal rule text in the final public copy (SKILL.md).\n- [COMMAND_EXECUTION]: The skill uses the dineway CLI tool to perform content creation and updates. These are recognized as legitimate vendor resources from the author 'dineway'.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 03:11 PM
Security Audit — agent-trust-hub — dineway-content-writer