research-review
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides setup instructions involving a
bashcommand to add an MCP server (claude mcp add codex ...). This is a standard configuration step for the Claude environment and targets a local script path within the user's home directory ($HOME/aris_repo). - [REMOTE_CODE_EXECUTION]: The skill references multiple external AI 'backends' (Codex, Antigravity, Grok). While these involve sending data to external services via MCP tools (e.g.,
mcp__codex__codex), the configuration and use cases are consistent with standard AI agent tool extensions and do not involve unauthorized remote code execution. - [DATA_EXPOSURE]: The skill instructs the agent to read project files (README.md, drafts, experimental notes) to build a research brief. This is the intended purpose of the skill and does not involve accessing sensitive system credentials or private keys outside the project scope.
Audit Metadata