research-hub-multi-ai

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local shell commands to check for the presence of the research-hub CLI (research-hub doctor) and to detect available LLM configurations via Python (python -c "..."). These are diagnostic checks used to ensure the execution environment is properly configured before generating plans.
  • [DATA_EXPOSURE]: The skill reads and writes to local project directories (.coord/, .ai/, .research/, .paper/). This is the intended purpose of the skill to manage research and coordination artifacts within a workspace.
  • [REMOTE_CODE_EXECUTION]: While the skill references other agents (codex-delegate, gemini-delegate), it does not download or execute arbitrary code from external sources. It acts as a router that generates structured instructions for other local tools to follow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 10:37 AM
Security Audit — agent-trust-hub — research-hub-multi-ai