CodeParse

CodeParse is a codebase indexer that gives agents fast, accurate, and compact context on your Python codebases while reducing token usage. It parses your codebase and stores the relationships of your code in a local SQLite database.

Install MCP Server and Skill

1. Put the codeparse binary on your PATH (release asset or make compile → dist/codeparse).
2. codeparse install

Benchmark Evaluation (in-progress)

Ten questions found in eval/tasks.json are evaluated against the SQLMesh repo using a baseline agent and an agent with the CodeParse MCP server. The average amount of tokens is calculated at the end for comparison.

Benchmark Legend

• AI Baseline: Builtin tools
• CodeParse: Builtin tools and CodeParse MCP Server w/ SKILL.md
• Tokens: Avg tokens used by the agent to complete a task.

Benchmark Results

Cursor Benchmark Evaluation - Grok 4.7
----------------- Mean -----------------
AI Baseline Tokene: 393,530
CodeParse Tokens: 265,867 (-32% reduction)
----------------------------------------

Claude Benchmark Evaluation - Opus 5.5 Medium
----------------- Mean -----------------
AI Baseline Tokene: XXXX
CodeParse Tokens: XXXX
----------------------------------------

Running the Evaluation

Cursor Default Model: grok 4.7
Claude Defualt Model: opus 5.5 medium

• make eval-smoke cursor will run all 10 questions and evaluate.
• make eval cursor will run all 10 questions on repeat (10 times) and evaluate.