Python · open-source evidence
Python repositories using AI coding agents.
Across 9,117 cataloged Python repositories, qmmit considered 1,380,703 commits and detected 188,004 agent-attributed commits—a 13.6% aggregate floor for this catalog slice.
Repository evidence
Python projects with the most detected agent commits.
Only observations with at least 100 considered commits appear here. This orders detected evidence; it does not score project quality.
BasedHardware/omi
AI that sees your screen, listens to your conversations and tells you what to do
- Detected AI share
- 41.9%
- agent commits
- 8,288
- Commits considered
- 19,768
PostHog/posthog
:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
- Detected AI share
- 16.7%
- agent commits
- 4,155
- Commits considered
- 24,854
spec-kitty/spec-kitty
Spec-Driven Development with organizational governance. Specs tell AI agents what to build; Charter governs how they build it. Git-native missions, enforceable workflows, and consistent engineering standards across AI coding agents and harnesses.
- Detected AI share
- 38.8%
- agent commits
- 3,636
- Commits considered
- 9,370
BerriAI/litellm
The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
- Detected AI share
- 23.2%
- agent commits
- 3,096
- Commits considered
- 13,360
- Detected AI share
- 24.3%
- agent commits
- 2,977
- Commits considered
- 12,261
tenstorrent/tt-metal
:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.
- Detected AI share
- 30.4%
- agent commits
- 2,391
- Commits considered
- 7,875
mlflow/mlflow
The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.
- Detected AI share
- 72.6%
- agent commits
- 2,046
- Commits considered
- 2,819
home-assistant/core
:house_with_garden: Open source home automation that puts local control and privacy first.
- Detected AI share
- 16.6%
- agent commits
- 2,024
- Commits considered
- 12,164
- Detected AI share
- 90.9%
- agent commits
- 1,962
- Commits considered
- 2,159
fim-ai/fim-one
Source-available agent platform for Global × China enterprises — wire every system through one agent core. Self-hosted, any LLM.
- Detected AI share
- 93%
- agent commits
- 1,652
- Commits considered
- 1,777
airbytehq/airbyte
Open-source data movement for ELT pipelines and AI agents — from APIs, databases & files to warehouses, lakes, and AI applications. Both self-hosted and Cloud.
- Detected AI share
- 87.9%
- agent commits
- 1,611
- Commits considered
- 1,832
openakita/openakita
An open-source AI assistant framework with skills and agent architecture
- Detected AI share
- 42.3%
- agent commits
- 1,554
- Commits considered
- 3,678
apache/superset
Apache Superset is a Data Visualization and Data Exploration Platform
- Detected AI share
- 53.7%
- agent commits
- 1,519
- Commits considered
- 2,831
- Detected AI share
- 53.8%
- agent commits
- 1,458
- Commits considered
- 2,711
HKUDS/LightRAG
[EMNLP2025] LightRAG: Simple and Fast Retrieval-Augmented Generation
- Detected AI share
- 43.9%
- agent commits
- 1,401
- Commits considered
- 3,188
noonghunna/club-3090
Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, sglang, llama.cpp and other custom engines) and model-agnostic. Currently shipping Qwen3.8-27B Qwen3.6 35B Gemma 4 26B Gemma 4 31B configs for 1×/2×/4x/8x cards.
- Detected AI share
- 79.1%
- agent commits
- 1,395
- Commits considered
- 1,763
jgravelle/jcodemunch-mcp
Symbol-level code retrieval for AI agents: fetch the function you need, not the whole file. Tree-sitter powered, ~95% fewer tokens per lookup. Works with Claude Code, Cursor, Codex & any MCP client. 1.3T+ tokens saved.
- Detected AI share
- 76.6%
- agent commits
- 1,384
- Commits considered
- 1,806
Open-Source-Legal/OpenContracts
The open document intelligence platform for builders and hackers - DMS for the agentic world
- Detected AI share
- 31.5%
- agent commits
- 1,274
- Commits considered
- 4,040
- Detected AI share
- 77.9%
- agent commits
- 1,247
- Commits considered
- 1,601
- Detected AI share
- 78.4%
- agent commits
- 1,161
- Commits considered
- 1,480
raullenchai/Rapid-MLX
Rapid-MLX is an open-source (Apache 2.0) OpenAI- and Anthropic-compatible LLM inference server for Apple Silicon, built on MLX, focused on reliable tool calling for coding agents. Up to 4× faster than Apple's MLX (mlx-lm) on the same weights.
- Detected AI share
- 48.6%
- agent commits
- 1,159
- Commits considered
- 2,386
teng-lin/notebooklm-py
Unofficial Python API and agentic skill for Google Gemini Notebook. Full programmatic access to Gemini Notebook's features—including capabilities the web UI doesn't expose—via Python, CLI, and AI agents like Claude Code, Codex, and OpenClaw.
- Detected AI share
- 61.6%
- agent commits
- 1,150
- Commits considered
- 1,866
microsoft/agent-governance-toolkit
AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10.
- Detected AI share
- 63.4%
- agent commits
- 1,141
- Commits considered
- 1,799
jhd3197/ServerKit
ServerKit is a lightweight, modern server control panel for managing web applications, databases, and services on your VPS or dedicated server. Built with Python/Flask backend and React frontend.
- Detected AI share
- 67.9%
- agent commits
- 1,117
- Commits considered
- 1,646
Use this ecosystem view carefully
GitHub language classification is a repository-level snapshot, while the detector reads commit metadata. The aggregate is descriptive of this catalog slice and is not the published benchmark.