Libraries and SDKs

Lightweight runtimes, tool loops, and provider-agnostic harness primitives.

#ProjectStarsTierOSSDescription
1Daytona
sandbox
72.4kslightly complexElastic dev environments for AI-generated code: workspaces, Git, previews—infra harness between "the model wrote a patch" and "it ran in a real machine."
2Mem0
memorypython
59kslightly complexUniversal memory layer for AI agents: stores user/org/session memory, retrieves on demand. Apache-2.0; the de-facto memory primitive paired with most harnesses in 2026.
3LiteLLM
provider-agnosticpython
51kmostly simpleOne interface to 100+ LLMs; routing, caching, budgets. Not an agent framework—the pipe every agent framework uses.
4Composio
sandboxtool-discoverypythontypescript
28.9kcomplex1,000+ toolkits with auth, tool search, and a sandboxed workbench—drop-in tool layer so agents stop reinventing OAuth + integrations. Python and TypeScript.
5smolagents
sandboxpython
27.9kmostly simpleCode-as-action agents: model outputs Python executed in sandbox (E2B, Modal, etc.); ~1k LOC core.
6vercel/ai
provider-agnostictypescript
25kslightly complexReact and Node SDK for streaming, tool calls, and agent-style UIs; provider-agnostic.
7deepagents
multi-agentsandboxpythontypescript
24.9kslightly complexLangChain's Python+TypeScript agent harness on top of LangGraph: planning tool, virtual filesystem, shell sandbox, sub-agent spawning—the "Claude Code-style" harness as a reusable library.
8pydantic-ai
mcptypedprovider-agnosticpython
17.9kslightly complexType-safe Python agents with Pydantic I/O; multi-provider, MCP, Logfire observability, and human-in-the-loop.
9E2B
sandboxpython
12.7kslightly complexFirecracker sandboxes for executing agent-generated code; the hosted isolation layer many tool-calling demos use instead of running arbitrary LLM output on your laptop.
10strands-agents
mcpmulti-agenttypedpython
6.2kmostly simpleModel-driven Python SDK; decorators for tools, native MCP, multi-agent; "minimal code" without sacrificing provider choice.
11Cloudflare Agents
memorytypescript
5.1kslightly complexPersistent, stateful agents on Durable Objects: state, websockets, scheduling, and AI chat baked in. The serverless answer to "where does the agent live?"
12openai-agents-js
multi-agentvoicetypescript
3.3kslightly complexOfficial OpenAI Agents SDK for Node/TS: handoffs, guardrails, voice; the JS counterpart to openai-agents-python.
13open-harness
mcpmulti-agenttypescript
571slightly complexTypeScript Agent class on Vercel AI SDK; streaming events, filesystem/bash tools, MCP, and subagent delegation.
14Community-curated agent lists
11super simpleBroader directories: e.g. [brandonhimpfen/awesome-ai-agents](https://github.com/brandonhimpfen/awesome-ai-agents), [axioma-ai-labs/awesome-ai-agent-frameworks](https://github.com/axioma-ai-labs/awesome-ai-agent-frameworks), [mb-mal/awesome-ai-agents-frameworks](https://github.com/mb-mal/awesome-ai-agents-frameworks)—differ by scope and update cadence.