{"generated":"2026-07-06T15:04:54.420Z","count":33,"tools":[{"slug":"mcp-servers","name":"MCP Servers","owner":"modelcontextprotocol","repo":"servers","category":"mcp","lang":"TypeScript","blurb":"The reference collection of Model Context Protocol servers — connect agents to files, GitHub, databases, and more.","use_cases":"[\"tool integration\",\"agent connectors\",\"MCP ecosystem\"]","alternatives":["fastmcp"],"stars":88103,"pushed_at":"2026-07-04T23:04:27Z","synced_at":"2026-07-06T09:27:45.999Z","useCases":["tool integration","agent connectors","MCP ecosystem"]},{"slug":"mem0","name":"Mem0","owner":"mem0ai","repo":"mem0","category":"memory","lang":"Python","blurb":"A memory layer for AI agents — extracts, stores, and retrieves user/agent facts across sessions.","use_cases":"[\"personalized agents\",\"cross-session memory\",\"user preference recall\"]","alternatives":["letta","zep"],"stars":60186,"pushed_at":"2026-07-06T09:27:33Z","synced_at":"2026-07-06T09:27:46.208Z","useCases":["personalized agents","cross-session memory","user preference recall"]},{"slug":"autogen","name":"AutoGen","owner":"microsoft","repo":"autogen","category":"framework","lang":"Python","blurb":"Microsoft's framework for multi-agent conversation, with a programming model for agents that talk to each other and tools.","use_cases":"[\"conversational multi-agent\",\"code execution\",\"research\"]","alternatives":["langgraph","crewai","semantic-kernel"],"stars":59522,"pushed_at":"2026-04-15T11:59:09Z","synced_at":"2026-07-06T09:27:46.356Z","useCases":["conversational multi-agent","code execution","research"]},{"slug":"crewai","name":"CrewAI","owner":"crewAIInc","repo":"crewAI","category":"framework","lang":"Python","blurb":"Role-playing autonomous agents that collaborate as a 'crew' with defined roles, goals, and task delegation.","use_cases":"[\"role-based multi-agent teams\",\"task delegation\",\"rapid prototyping\"]","alternatives":["langgraph","autogen"],"stars":54995,"pushed_at":"2026-07-05T23:27:44Z","synced_at":"2026-07-06T09:27:46.502Z","useCases":["role-based multi-agent teams","task delegation","rapid prototyping"]},{"slug":"llama-index","name":"LlamaIndex","owner":"run-llama","repo":"llama_index","category":"framework","lang":"Python","blurb":"Data framework for connecting LLMs to private data — indexing, retrieval, and agentic RAG over your documents.","use_cases":"[\"RAG\",\"document agents\",\"knowledge assistants\"]","alternatives":["langgraph","haystack"],"stars":50679,"pushed_at":"2026-07-02T17:54:20Z","synced_at":"2026-07-06T09:27:46.626Z","useCases":["RAG","document agents","knowledge assistants"]},{"slug":"milvus","name":"Milvus","owner":"milvus-io","repo":"milvus","category":"vectordb","lang":"Go","blurb":"Cloud-native vector database built for billion-scale similarity search.","use_cases":"[\"billion-scale search\",\"enterprise RAG\",\"distributed\"]","alternatives":["qdrant","weaviate"],"stars":45090,"pushed_at":"2026-07-06T09:26:37Z","synced_at":"2026-07-06T09:27:46.773Z","useCases":["billion-scale search","enterprise RAG","distributed"]},{"slug":"agno","name":"Agno","owner":"agno-agi","repo":"agno","category":"framework","lang":"Python","blurb":"Batteries-included agent runtime — built-in memory, knowledge/RAG, and AgentOS, a control plane you run in your own cloud. Formerly Phidata.","use_cases":"[\"full-stack agents\",\"built-in memory + RAG\",\"self-hosted control plane\"]","alternatives":["langgraph","crewai","pydantic-ai"],"stars":41012,"pushed_at":"2026-07-06T09:02:22Z","synced_at":"2026-07-06T09:27:46.890Z","useCases":["full-stack agents","built-in memory + RAG","self-hosted control plane"]},{"slug":"duckdb","name":"DuckDB","owner":"duckdb","repo":"duckdb","category":"vectordb","lang":"C++","blurb":"In-process analytical database whose vss extension adds an HNSW vector index — vector search alongside your columnar analytics.","use_cases":"[\"analytical + vector search\",\"embedded OLAP\",\"batch retrieval\"]","alternatives":["lancedb","sqlite-vec","pgvector"],"stars":39202,"pushed_at":"2026-07-06T08:28:10Z","synced_at":"2026-07-06T09:27:47.043Z","useCases":["analytical + vector search","embedded OLAP","batch retrieval"]},{"slug":"langgraph","name":"LangGraph","owner":"langchain-ai","repo":"langgraph","category":"framework","lang":"Python","blurb":"Graph-based orchestration for stateful, multi-actor agent workflows with explicit control flow and checkpointing.","use_cases":"[\"stateful multi-agent workflows\",\"human-in-the-loop\",\"long-running agents\"]","alternatives":["crewai","autogen","llama-index"],"stars":36609,"pushed_at":"2026-07-05T00:59:26Z","synced_at":"2026-07-06T09:27:47.173Z","useCases":["stateful multi-agent workflows","human-in-the-loop","long-running agents"]},{"slug":"dspy","name":"DSPy","owner":"stanfordnlp","repo":"dspy","category":"framework","lang":"Python","blurb":"Programming — not prompting — language models: compile declarative pipelines into optimized prompts/weights.","use_cases":"[\"prompt optimization\",\"compound AI systems\",\"self-improving pipelines\"]","alternatives":["langgraph","llama-index"],"stars":35864,"pushed_at":"2026-07-05T03:46:11Z","synced_at":"2026-07-06T09:27:47.285Z","useCases":["prompt optimization","compound AI systems","self-improving pipelines"]},{"slug":"qdrant","name":"Qdrant","owner":"qdrant","repo":"qdrant","category":"vectordb","lang":"Rust","blurb":"High-performance vector search engine with rich filtering, written in Rust for production-scale retrieval.","use_cases":"[\"production RAG\",\"filtered search\",\"high throughput\"]","alternatives":["chroma","weaviate","milvus"],"stars":32969,"pushed_at":"2026-07-06T08:23:29Z","synced_at":"2026-07-06T09:27:47.412Z","useCases":["production RAG","filtered search","high throughput"]},{"slug":"langfuse","name":"Langfuse","owner":"langfuse","repo":"langfuse","category":"observability","lang":"TypeScript","blurb":"Open-source LLM engineering platform — tracing, evals, prompt management, and metrics for agent apps.","use_cases":"[\"LLM tracing\",\"prompt management\",\"production monitoring\"]","alternatives":["phoenix","helicone"],"stars":30515,"pushed_at":"2026-07-06T09:04:20Z","synced_at":"2026-07-06T09:27:47.542Z","useCases":["LLM tracing","prompt management","production monitoring"]},{"slug":"chroma","name":"Chroma","owner":"chroma-core","repo":"chroma","category":"vectordb","lang":"Rust","blurb":"Open-source embedding database designed for simplicity — the default vector store for many RAG prototypes.","use_cases":"[\"RAG\",\"semantic search\",\"local prototyping\"]","alternatives":["qdrant","weaviate","milvus"],"stars":28706,"pushed_at":"2026-07-06T07:31:27Z","synced_at":"2026-07-06T09:27:47.653Z","useCases":["RAG","semantic search","local prototyping"]},{"slug":"openai-agents-sdk","name":"OpenAI Agents SDK","owner":"openai","repo":"openai-agents-python","category":"framework","lang":"Python","blurb":"OpenAI's lightweight agent framework — a small set of primitives (Agents, Handoffs, Guardrails, Sessions); provider-agnostic via LiteLLM. Evolved from Swarm.","use_cases":"[\"minimal orchestration\",\"agent handoffs\",\"guardrailed agents\"]","alternatives":["pydantic-ai","agno","langgraph"],"stars":27685,"pushed_at":"2026-07-06T06:41:40Z","synced_at":"2026-07-06T09:27:47.780Z","useCases":["minimal orchestration","agent handoffs","guardrailed agents"]},{"slug":"fastmcp","name":"FastMCP","owner":"jlowin","repo":"fastmcp","category":"mcp","lang":"Python","blurb":"The fast, Pythonic way to build MCP servers and clients — decorators over boilerplate.","use_cases":"[\"building MCP servers\",\"Python tooling\",\"rapid integration\"]","alternatives":["mcp-servers"],"stars":25991,"pushed_at":"2026-07-05T23:28:08Z","synced_at":"2026-07-06T09:27:48.001Z","useCases":["building MCP servers","Python tooling","rapid integration"]},{"slug":"letta","name":"Letta (MemGPT)","owner":"letta-ai","repo":"letta","category":"memory","lang":"Python","blurb":"Stateful agents with long-term memory and self-editing context, evolved from the MemGPT research.","use_cases":"[\"persistent agents\",\"self-editing memory\",\"long conversations\"]","alternatives":["mem0","zep"],"stars":23662,"pushed_at":"2026-07-03T18:53:41Z","synced_at":"2026-07-06T09:27:48.115Z","useCases":["persistent agents","self-editing memory","long conversations"]},{"slug":"promptfoo","name":"promptfoo","owner":"promptfoo","repo":"promptfoo","category":"eval","lang":"TypeScript","blurb":"Test-driven prompt and agent development — evals, red-teaming, and side-by-side model comparison from the CLI.","use_cases":"[\"prompt evals\",\"red-teaming\",\"model comparison\"]","alternatives":["deepeval","ragas"],"stars":22958,"pushed_at":"2026-07-06T08:20:28Z","synced_at":"2026-07-06T09:27:48.234Z","useCases":["prompt evals","red-teaming","model comparison"]},{"slug":"pgvector","name":"pgvector","owner":"pgvector","repo":"pgvector","category":"vectordb","lang":"C","blurb":"Vector similarity search inside Postgres — keep embeddings next to your relational data.","use_cases":"[\"RAG on existing Postgres\",\"no new infra\",\"transactional + vector\"]","alternatives":["qdrant","chroma"],"stars":22081,"pushed_at":"2026-07-02T00:21:28Z","synced_at":"2026-07-06T09:27:48.408Z","useCases":["RAG on existing Postgres","no new infra","transactional + vector"]},{"slug":"temporal","name":"Temporal","owner":"temporalio","repo":"temporal","category":"runtime","lang":"Go","blurb":"Durable execution platform — write long-running, failure-resilient agent workflows as ordinary code.","use_cases":"[\"durable agent workflows\",\"retries & recovery\",\"long-running orchestration\"]","alternatives":["inngest"],"stars":21440,"pushed_at":"2026-07-06T05:34:20Z","synced_at":"2026-07-06T09:27:48.524Z","useCases":["durable agent workflows","retries & recovery","long-running orchestration"]},{"slug":"google-adk","name":"Google ADK","owner":"google","repo":"adk-python","category":"framework","lang":"Python","blurb":"Google's Agent Development Kit — a code-first, model-agnostic toolkit for building, evaluating, and deploying multi-agent systems. Optimized for Gemini.","use_cases":"[\"multi-agent systems\",\"agent evaluation\",\"Gemini-first agents\"]","alternatives":["langgraph","crewai","openai-agents-sdk"],"stars":20490,"pushed_at":"2026-07-06T07:44:41Z","synced_at":"2026-07-06T09:27:48.648Z","useCases":["multi-agent systems","agent evaluation","Gemini-first agents"]},{"slug":"pydantic-ai","name":"Pydantic AI","owner":"pydantic","repo":"pydantic-ai","category":"framework","lang":"Python","blurb":"Type-safe agent framework from the Pydantic team — structured outputs, dependency injection, and model-agnostic agents.","use_cases":"[\"type-safe agents\",\"structured outputs\",\"production Python\"]","alternatives":["langgraph","crewai"],"stars":18245,"pushed_at":"2026-07-06T09:12:54Z","synced_at":"2026-07-06T09:27:48.777Z","useCases":["type-safe agents","structured outputs","production Python"]},{"slug":"deepeval","name":"DeepEval","owner":"confident-ai","repo":"deepeval","category":"eval","lang":"Python","blurb":"Pytest-like framework for unit-testing LLM outputs with metrics for hallucination, relevancy, and bias.","use_cases":"[\"LLM unit tests\",\"regression testing\",\"CI gates\"]","alternatives":["ragas","promptfoo"],"stars":16665,"pushed_at":"2026-07-02T18:41:12Z","synced_at":"2026-07-06T09:27:48.892Z","useCases":["LLM unit tests","regression testing","CI gates"]},{"slug":"weaviate","name":"Weaviate","owner":"weaviate","repo":"weaviate","category":"vectordb","lang":"Go","blurb":"Open-source vector database with hybrid search and built-in modules for vectorization and RAG.","use_cases":"[\"hybrid search\",\"RAG\",\"multimodal\"]","alternatives":["qdrant","chroma","milvus"],"stars":16515,"pushed_at":"2026-07-06T09:17:57Z","synced_at":"2026-07-06T09:27:49.002Z","useCases":["hybrid search","RAG","multimodal"]},{"slug":"ragas","name":"Ragas","owner":"explodinggradients","repo":"ragas","category":"eval","lang":"Python","blurb":"Evaluation toolkit for RAG pipelines — faithfulness, answer relevancy, and context metrics without ground truth.","use_cases":"[\"RAG evaluation\",\"faithfulness scoring\",\"CI for retrieval\"]","alternatives":["deepeval","promptfoo"],"stars":14675,"pushed_at":"2026-02-24T07:47:19Z","synced_at":"2026-07-06T09:27:49.194Z","useCases":["RAG evaluation","faithfulness scoring","CI for retrieval"]},{"slug":"e2b","name":"E2B","owner":"e2b-dev","repo":"E2B","category":"runtime","lang":"TypeScript","blurb":"Secure cloud sandboxes for running AI-generated code — the runtime layer for code-executing agents.","use_cases":"[\"code execution\",\"agent sandboxes\",\"data analysis agents\"]","alternatives":["modal"],"stars":12854,"pushed_at":"2026-07-03T23:54:24Z","synced_at":"2026-07-06T09:27:49.319Z","useCases":["code execution","agent sandboxes","data analysis agents"]},{"slug":"lancedb","name":"LanceDB","owner":"lancedb","repo":"lancedb","category":"vectordb","lang":"Rust","blurb":"Embedded, in-process vector database on the columnar Lance format — versioned, updatable, larger-than-RAM retrieval with no server.","use_cases":"[\"embedded vector search\",\"mutable/versioned datasets\",\"multimodal RAG\"]","alternatives":["sqlite-vec","chroma","duckdb"],"stars":10806,"pushed_at":"2026-07-04T18:11:12Z","synced_at":"2026-07-06T09:27:49.442Z","useCases":["embedded vector search","mutable/versioned datasets","multimodal RAG"]},{"slug":"phoenix","name":"Phoenix","owner":"Arize-ai","repo":"phoenix","category":"observability","lang":"Jupyter Notebook","blurb":"Arize's open-source observability for LLM apps — OpenTelemetry-based tracing and evaluation.","use_cases":"[\"OTel tracing\",\"evaluation\",\"debugging RAG\"]","alternatives":["langfuse","helicone"],"stars":10414,"pushed_at":"2026-07-06T05:33:45Z","synced_at":"2026-07-06T09:27:49.575Z","useCases":["OTel tracing","evaluation","debugging RAG"]},{"slug":"sqlite-vec","name":"sqlite-vec","owner":"asg017","repo":"sqlite-vec","category":"vectordb","lang":"C","blurb":"A single-file SQLite extension for vector search — exact brute-force KNN that lives inside the database you already ship.","use_cases":"[\"vectors inside SQLite\",\"edge/local agents\",\"exact small-scale search\"]","alternatives":["lancedb","pgvector","chroma"],"stars":7828,"pushed_at":"2026-05-18T06:50:45Z","synced_at":"2026-07-06T09:27:49.676Z","useCases":["vectors inside SQLite","edge/local agents","exact small-scale search"]},{"slug":"claude-agent-sdk","name":"Claude Agent SDK","owner":"anthropics","repo":"claude-agent-sdk-python","category":"framework","lang":"Python","blurb":"Anthropic's SDK for building agents on Claude — the harness behind Claude Code, with in-process tools, hooks, and an interactive client. Formerly the Claude Code SDK.","use_cases":"[\"Claude-native agents\",\"coding agents\",\"custom tools + hooks\"]","alternatives":["langgraph","openai-agents-sdk","pydantic-ai"],"stars":7533,"pushed_at":"2026-07-06T00:04:48Z","synced_at":"2026-07-06T09:27:49.822Z","useCases":["Claude-native agents","coding agents","custom tools + hooks"]},{"slug":"strands-agents","name":"Strands Agents","owner":"strands-agents","repo":"sdk-python","category":"framework","lang":"Python","blurb":"AWS's model-driven agent SDK — give it a prompt and tools and let the model plan, call tools, and reflect in a loop. First-class Bedrock, MCP-native, OpenTelemetry tracing.","use_cases":"[\"AWS/Bedrock agents\",\"model-driven loops\",\"fast prototyping\"]","alternatives":["langgraph","google-adk","openai-agents-sdk"],"stars":6431,"pushed_at":"2026-07-06T07:57:03Z","synced_at":"2026-07-06T09:27:50.095Z","useCases":["AWS/Bedrock agents","model-driven loops","fast prototyping"]},{"slug":"helicone","name":"Helicone","owner":"Helicone","repo":"helicone","category":"observability","lang":"TypeScript","blurb":"Open-source observability for LLM apps via a proxy — logging, caching, and cost tracking with one header.","use_cases":"[\"cost tracking\",\"request logging\",\"caching\"]","alternatives":["langfuse","phoenix"],"stars":5913,"pushed_at":"2026-07-05T05:55:34Z","synced_at":"2026-07-06T09:27:50.206Z","useCases":["cost tracking","request logging","caching"]},{"slug":"cloudflare-agents","name":"Cloudflare Agents","owner":"cloudflare","repo":"agents","category":"framework","lang":"TypeScript","blurb":"TypeScript-first SDK for stateful agents where each agent is a Durable Object — embedded SQLite, WebSockets, cron scheduling, and hibernation, running globally on Cloudflare's edge.","use_cases":"[\"edge AI agents\",\"stateful serverless agents\",\"one agent per user\"]","alternatives":["langgraph","claude-agent-sdk"],"stars":5228,"pushed_at":"2026-07-03T22:33:13Z","synced_at":"2026-07-06T09:27:50.359Z","useCases":["edge AI agents","stateful serverless agents","one agent per user"]},{"slug":"zep","name":"Zep","owner":"getzep","repo":"zep","category":"memory","lang":"Go","blurb":"Long-term memory store for agents with a temporal knowledge graph of facts and their validity over time.","use_cases":"[\"temporal memory\",\"knowledge graphs\",\"fact recall\"]","alternatives":["mem0","letta"],"stars":4730,"pushed_at":"2026-07-03T21:21:39Z","synced_at":"2026-07-06T09:27:50.572Z","useCases":["temporal memory","knowledge graphs","fact recall"]}]}