storm-research-agent

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a highly structured STORM-style research workflow that emphasizes multi-perspective analysis (Practitioner, Academic, Skeptic, Economist, Historian) and evidence-based decision making.
  • [SAFE]: It defines clear operational boundaries in the SKILL.md and output-contract.md, explicitly instructing the agent to avoid downstream actions such as publishing to social media, creating drafts, or sending messages without user consent.
  • [SAFE]: Data management is handled safely, with instructions to save research outputs to a specific local directory (03_analysis/) and labels for claims based on their source reliability (sourced, inferred, needs verification).
  • [SAFE]: No malicious patterns such as prompt injection, obfuscation, hardcoded credentials, or remote code execution were detected across any of the analyzed skill files.
  • [SAFE]: The skill includes extensive self-verification and risk-profiling reports (output-risk-profile.md, package-verification.md) which suggest a focus on quality control and safety best practices during development.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 07:27 PM
Security Audit — agent-trust-hub — storm-research-agent