johnnichev/selectools

Production-ready Python framework for AI agents with built-in guardrails, audit logging, cost tracking, and hybrid RAG. Supports OpenAI, Anthropic, Gemini, Ollama. By NichevLabs.

43
/ 100
Emerging

Implements multi-agent orchestration and composable pipelines natively in Python—AgentGraph supports directed routing, four supervisor strategies (plan-and-execute, round-robin, dynamic, Magentic-One), and human-in-the-loop interrupts that resume at exact yield points rather than restarting nodes. Includes semantic caching, prompt compression, token estimation, cost budgeting, parallel execution with merge policies, checkpointing across three backends, and 39 built-in evaluators for agent testing without external services.

6 stars and 4,332 monthly downloads. Available on PyPI.

Maintenance 13 / 25
Adoption 12 / 25
Maturity 18 / 25
Community 0 / 25

How are scores calculated?

Stars

6

Forks

Language

Python

License

Apache-2.0

Last pushed

Mar 28, 2026

Monthly downloads

4,332

Commits (30d)

0

Dependencies

4

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/johnnichev/selectools"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.