Agent Claude Code
An architecture review agent that examines how a project's modules, layers, dependencies, and interchangeable backends fit together. It reports issues with file-and-line evidence but does not edit files.
Agent Claude Code
A code review agent that runs the project's linter, tests, and dry evaluation before examining the implementation. It reports code and test problems with file-and-line evidence but does not edit files.
Agent Claude Code
A documentation review agent that checks a project's README and documentation against its code. It reports problems but does not edit files.
Agent Claude Code
A pre-demo stability checker that runs the project's command-line entry points and failure scenarios using mock data. It reports what works and what fails but does not edit files.
Hook Claude Code
Runs before the agent uses a tool for Bash tool calls, executing block-live-commands.py via python3. From Trum-ok/tutu-mcp-hackathon.
Settings file Claude Code
Agent settings declaring 1 hook event (PreToolUse).
Instructions file
A project instruction file for Claude Code, an AI coding assistant. It sets rules for Russian documentation, comments, source-of-truth files, Git practices, and blocked live-mode commands.
MCP server Claude CodeCodexCursor +2
MCP server "tutu-mcp" as configured in Trum-ok/tutu-mcp-hackathon. Runs locally from the tutu-mcp Python package.