Skip to main content

CodeMind OKF CLI — Generate AI-ready knowledge bundles for any codebase. Works with Cursor, Antigravity, GitHub Copilot, and any MCP-compatible AI IDE.

Project description

🧠 CodeMind OKF CLI

Generate AI-ready knowledge bundles for any codebase.
Works with Cursor, Antigravity IDE, Claude Desktop, GitHub Copilot, and any MCP-compatible AI IDE.

PyPI Version Python Versions MCP Compatible


📦 Installation

pip install codemind-okf

🚀 Quick Usage

1. codemind index — Index Any Project

Crawls your project files using Tree-sitter & AST parsers to generate a .okf/ knowledge bundle.

codemind index .                         # Index current directory
codemind index /path/to/project          # Index a specific directory
codemind index . --lang python           # Only index Python files
codemind index . --overwrite             # Force re-index from scratch
  • ⚡ SHA-256 Incremental Indexing: Remembers file hashes. Subsequent runs skip unchanged files in 0.05s!

2. codemind init — Drop AI IDE Instructions

Creates instruction files so Cursor, Antigravity, and GitHub Copilot use .okf/index.md as primary context.

codemind init                            # Create all AI IDE config files

Files created:

File Compatible Tool
.cursorrules Cursor AI IDE
.agents/AGENTS.md Antigravity IDE & Open Agents
.github/copilot-instructions.md GitHub Copilot

3. codemind audit — Architecture & Health Audit

Performs an architectural health check evaluating line density, docstring coverage, and layer separation.

codemind audit                          # Health score (0-100 A+)

4. codemind watch — Real-Time Background File Watcher

Watches the project directory for file saves and incrementally updates .okf/ in under 50ms.

codemind watch                          # Run watcher in background

5. codemind mcp — Model Context Protocol (MCP) Server

Launches the MCP server over STDIO. Connects CodeMind natively to Cursor, Claude Desktop, Antigravity, or Zed.

codemind mcp

Add to Cursor / Claude Desktop / Antigravity (mcp.json):

{
  "mcpServers": {
    "codemind": {
      "command": "uvx",
      "args": [
        "--from",
        "codemind-okf",
        "codemind",
        "mcp"
      ]
    }
  }
}

6. codemind memory — Manage AI Persistent Memory Log (.okf/memory.md)

Inspect, view, clear, or manually log memory entries to .okf/memory.md.

codemind memory show                     # Print full memory log
codemind memory ls                       # Show entry summary by type
codemind memory add "..." --type decision# Manually append memory entry
codemind memory clear                    # Reset memory log

🔌 Exposed MCP Tools

MCP Tool Description
🧠 remember(content, type) Persists AI decisions, tasks, context & bugs into .okf/memory.md
🔄 recall(query) Retrieves relevant past memories at session start to eliminate context loss
📚 get_project_index Reads the master architecture map (index.md)
🔍 search_bundle(query) Sub-word & token search across all project modules
📖 read_module(slug) Reads specific module AST metadata & function signatures
🔗 trace_dependencies(slug) Traces import graphs between components
📁 list_bundles Discovers available .okf/ knowledge bundles

📄 License

MIT License. See repository for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

codemind_okf-1.1.0.tar.gz (30.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

codemind_okf-1.1.0-py3-none-any.whl (33.8 kB view details)

Uploaded Python 3

File details

Details for the file codemind_okf-1.1.0.tar.gz.

File metadata

  • Download URL: codemind_okf-1.1.0.tar.gz
  • Upload date:
  • Size: 30.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for codemind_okf-1.1.0.tar.gz
Algorithm Hash digest
SHA256 a313b77a7d203a4cc15274c86aa5988a5889dcdd3521786494b113847925b8d6
MD5 43fcba813b89ecb96b3e9d65e51eb14f
BLAKE2b-256 437c6f630645e16e601944f145e28059cbc391d94e936993184f17ea3b345e66

See more details on using hashes here.

File details

Details for the file codemind_okf-1.1.0-py3-none-any.whl.

File metadata

  • Download URL: codemind_okf-1.1.0-py3-none-any.whl
  • Upload date:
  • Size: 33.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for codemind_okf-1.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 52d7ccc1e1d5660787ddb48e8879a574b63c0a6ca0c1fc86bc3af4704a5fb797
MD5 6c01292494e7e20f264866db918d53f2
BLAKE2b-256 1137b479addfdf58097f4bf1856a1f245bc067156af531f5c68aa2c76a36764a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page