# vstorm-co/pydantic-deepagents

Open-source, self-hosted Claude Code - a terminal AI assistant and the Python framework behind it. Tool-calling, sandboxed execution, multi-agent teams, skills, checkpoints, unlimited context - on Pydantic AI, any model.

Repository: https://github.com/vstorm-co/pydantic-deepagents
Canonical: https://ross.abutalabs.com/products/pydantic-deepagents
Homepage: https://vstorm-co.github.io/pydantic-deepagents/
Language: Python
License: MIT
License Family: permissive
Topics: pydantic-ai, agent-framework, pydantic, python, anthropic, llms, mcp, ai-agents, subagents, claude-code, cli, coding-agent, deep-research, docker-sandbox, playwright, tui, vstorm
Last push: 2026-08-22T14:57:43+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 84, longevity 19
- inputs: {"age_days": 277, "days_push": 11, "days_rel": 28, "gap_med": 1, "n_releases_24m": 66}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1044, forks 127 (observed 2026-08-28T04:03:21.224356+00:00)

## What it is
Pydantic Deep Agents is an open-source, self-hosted terminal AI assistant (a Claude Code alternative) plus the Python framework behind it, built on Pydantic AI. It provides tool-calling, sandboxed execution, multi-agent teams, memory, checkpoints, and unlimited context handling with any supported LLM.

## Use cases
- self-hosted open-source alternative to Claude Code
- build an AI coding agent in Python
- run a terminal AI assistant with any LLM
- orchestrate multi-agent teams with subagents
- execute agent code in a sandboxed Docker environment
- handle conversations longer than the model context window
- add MCP tools and web browsing to an AI agent

## When to choose
- you want a self-hosted, model-agnostic Claude Code alternative
- you're building agents in Python and want type-safe, batteries-included tooling on Pydantic AI
- you need filesystem, shell, web, memory, and sub-agents out of the box
- you need sandboxed execution and human-in-the-loop approval gates

## When to avoid
- you need a polished commercial coding assistant with vendor support
- you want a no-code or GUI-only agent builder
- you're not working in Python or don't want a framework dependency

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, cli, mcp, chatbot, developer-tools, terminal-ui
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cli, cross-platform, self-hosted
- tags: claude-code-alternative, pydantic-ai, coding-agent, subagents, docker-sandbox, deep-research, tui, tool-calling, checkpoints, unlimited-context, ai-agents, command-line, automation, docker

## Member repositories
- vstorm-co/pydantic-deepagents (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:21.224356+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:02:16.252214+00:00, confidence not recorded.
  - readme: https://github.com/vstorm-co/pydantic-deepagents (fetched 2026-08-28T04:03:21.224356+00:00, sha 75ce529f0867)
  - homepage: https://vstorm-co.github.io/pydantic-deepagents/ (fetched 2026-08-29T13:03:34.732498+00:00, sha e7c58388fe78)
- Data as of 2026-08-30T08:39:29.467469+00:00.