story-collaborator

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for creative writing collaboration and does not demonstrate any malicious intent or unauthorized capabilities.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill is restricted to local operations. It does not perform network requests or access sensitive system directories. File access is limited to project-specific paths like context/output-config.md and .story-collaborator-output.md for session persistence.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text to generate writing drafts, which is a standard feature for this use case.
  • Ingestion points: User-provided story drafts and collaboration directions (SKILL.md).
  • Boundary markers: No explicit delimiters are defined for user-provided drafts.
  • Capability inventory: Local file-write access to persist markdown content and configuration data within the project directory.
  • Sanitization: No specific sanitization or validation of user-provided content is mentioned before file persistence.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 05:34 PM