Back to Apps

Compare MCP
by cristophereasygoing927
Multi-model code review engine for Claude Code with ranked todos and subagent dispatch.
0 stars
Works in:claude
Exposes:Tools
What it does
Expands Claude Code's capabilities by fanning out code reviews to multiple LLMs simultaneously. It diffs insights from different models, calculates agreement rates, and manages a ranked todo list of fixes.
Tools
compare_run: Fan out code review to all enabled providers in parallel.compare_diff: Extract unique vs shared insights using fuzzy deduplication.compare_debate: Run a critique round where models challenge each other's findings.compare_todos: Persist ranked findings to a local SQLite store.compare_status: Monitor the state of pending and completed fix todos.compare_todo_update: Update the status of specific fix items.
Installation
Install via pip:
pip install compare-mcp
claude mcp add -s user compare-mcp -- python -m compare_mcp
Configure providers in ~/.compare/config.json.
Supported hosts
- Claude
Quick install
pip install compare-mcpInformation
- Pricing
- free
- Published
- 5/27/2026
- stars
- 0
Categories
Choose your AI client and follow the steps below.
Claude Desktop
pip install compare-mcp && claude mcp add -s user compare-mcp -- python -m compare_mcp





