scientific-writing-story

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill maps the 'research cloud' by reading various repository files which could potentially contain untrusted data.
  • Ingestion points: Repository files including experiments/scripts, results/, figures/, notebooks, git log, and CLAUDE.md in the consuming repository.
  • Boundary markers: None explicitly defined for the file reading process.
  • Capability inventory: The skill uses Explore subagents with read-only permissions and the superpowers:dispatching-parallel-agents tool to inventory the file system. It generates a markdown file (story.md).
  • Sanitization: No specific sanitization or filtering of file content is described before processing.
  • Note: The risk is significantly mitigated by the 'Red Flags' section and the procedural requirement to converge with the user and obtain explicit confirmation for the thesis and data split before generating the final output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 09:59 PM
Security Audit — agent-trust-hub — scientific-writing-story