Tool Use & Function Calling
Letting models call tools, emit structured output and run code. As of 2026-11-21, 275 indexed products match this subcategory (133 with full profiles); 90 were published in the last 90 days, against 54 in the 90 days before, slower than the catalogue as a whole. By GitHub stars, Daytona leads, holding 15% of the stars of the 235 products here with their own repository. The most common form is MCP servers (75% of indexed products).
SubcategorySegment label: Thin DER
Segment statistics
Whole catalogue (all 275 indexed products, not only profiled ones). Calculated 2026-11-21.
Thin. Among the smallest 30% of nodes at its level, and launches are not outpacing the catalogue. Size is ranked against the other subcategory nodes; momentum compares this node’s launches in the last 90 days with the 90 before, relative to the whole catalogue’s ratio (score 62 or more = faster, 38 or less = slower; below 5 dated launches it is not measured and the label uses size alone).
Concentration. The Herfindahl–Hirschman index (HHI) adds up each project’s squared share of stars. At 0.055, attention is spread as if across about 18 equally popular projects. Using the common competition-policy thresholds (0.15 and 0.25), that is unconcentrated. Stars measure attention, not usage or quality.
Which product forms serve it
The same need is met by MCP servers, agent skills, plugins, extensions and frameworks. This counts each form across the whole catalogue.
| Product type | Indexed | Share | With a profile |
|---|---|---|---|
| MCP servers | 206 | 74.9% | 67 |
| Frameworks | 20 | 7.3% | 18 |
| Infrastructure | 14 | 5.1% | 14 |
| Agent projects | 9 | 3.3% | 8 |
| Agent skills | 8 | 2.9% | 8 |
| Libraries | 4 | 1.5% | 4 |
| AI APIs | 4 | 1.5% | 4 |
| Plugins | 3 | 1.1% | 3 |
| IDE extensions | 3 | 1.1% | 3 |
| Agent SDKs | 1 | 0.36% | 1 |
| SaaS products | 1 | 0.36% | 1 |
| Rules packs | 1 | 0.36% | 1 |
| Platforms | 1 | 0.36% | 1 |
3 capabilities
Indexed = whole catalogue; profiled = products with their own page. A product can sit in several, so the rows do not add up to the total.
| Capability | Indexed | Profiled | New, 90 days | Label |
|---|---|---|---|---|
| MCP ClientsHosts and libraries that connect agents to MCP servers. | 75 | 21 | 26 | Lagging the catalogue |
| Structured OutputsGetting schema-valid JSON from models. | 60 | 26 | 23 | Established |
| Code Execution SandboxesIsolated environments where agents run code safely. | 45 | 20 | 19 | Lagging the catalogue |
Top 40 products by GitHub stars
Search the directoryProfiled products only (133 of 275), ranked by the stars of their own repository. Products that live in a subfolder of a larger repository (common for skills) show that whole repository's stars and are ranked after them; products without a repository come last. Stars are popularity, not quality or installs.
| # | Product | Type | Publisher | Capabilities | GitHub stars | Last push | Licence | Save and compare |
|---|---|---|---|---|---|---|---|---|
| 1 | DaytonaSandbox infrastructure that starts isolated environments for running AI-generated code quickly, with SDKs for files, git and processes. | Infrastructure | Daytona | 71k | 4mo ago | — | ||
| 2 | AgentScopePython multi-agent framework with message-based agent communication, tool use, memory and a visual studio for debugging. | Framework | Alibaba (AgentScope) | 36k | 1d ago | Apache-2.0 | ||
| 3 | OpenAI APIAPI for GPT and reasoning models via the Responses and Chat Completions endpoints, with built-in tools such as web search, file search, code interpreter and computer use. | API | openai | 32k | 1d ago | Apache-2.0 | ||
| 4 | ComposioIntegration platform with managed OAuth and thousands of tools across apps, exposed to agent frameworks via SDKs and MCP. | SaaS product | Composio | 31k | today | MIT | ||
| 5 | smolagentsMinimal Python library for agents that write and execute Python code as their actions, with sandboxed execution options and Hub integration. | Library | Hugging Face | 30k | 1d ago | Apache-2.0 | ||
| 6 | Semantic KernelSDK for integrating LLMs into .NET, Python and Java apps through plugins, function calling and agent abstractions. | Framework | microsoft | 29k | 1d ago | MIT | ||
| 7 | AI SDKTypeScript toolkit providing a unified API across model providers for text generation, structured output, tool calling and agent loops, plus React/Svelte/Vue UI hooks for streaming chat. | SDK | vercel | 28k | today | Apache-2.0 | ||
| 8 | HaystackPython framework for composing pipelines and agents from components for retrieval, generation and tool use, aimed at production RAG and search. | Framework | deepset | 27k | today | Apache-2.0 | ||
| 9 | E2BFirecracker microVM sandboxes that agents use to run untrusted code, with SDKs for files, processes and a code interpreter. | Infrastructure | E2B | 15k | today | Apache-2.0 | ||
| 10 | langchain4jLangChain4j is an idiomatic, open-source Java library for building LLM-powered applications on the JVM. It offers a unified API over popular LLM providers and vector stores, and makes implementing tool calling (including MCP support), agents and RAG easy. It integrates seamlessly with enterprise Java frameworks like Quarkus and Spring Boot. | Infrastructure | langchain4j | 13k | today | Apache-2.0 | ||
| 11 | mcp-useOpen-source library for connecting any LLM to MCP servers and building custom MCP agents and servers in Python or TypeScript. | Library | Manufact | 11k | today | MIT | ||
| 12 | aichatAll-in-one LLM CLI tool featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents, with access to OpenAI, Claude, Gemini, Ollama, Groq, and more. | Agent project | sigoden | 11k | 9mo ago | Apache-2.0 | ||
| 13 | mcpOpen source MCP Servers for AWS | MCP server | awslabs | 9.9k | today | Apache-2.0 | ||
| 14 | oumiEasily fine-tune (SFT/RL), evaluate, and deploy Qwen, Gemma, or any open agentic LLM/VLM! | Infrastructure | oumi-ai | 9.4k | today | Apache-2.0 | ||
| 15 | Awesome-MCP-ZHMCP 资源精选, MCP指南,Claude MCP,MCP Servers, MCP Clients | MCP server | yzfly | 7.9k | 7d ago | MIT | ||
| 16 | unstractLLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows | MCP server | Zipstack | 7.5k | 1d ago | AGPL-3.0 | ||
| 17 | smolvmAn embeddable, portable, branchable virtual machine to safely run Agents locally. | Agent project | smol-machines | 7.2k | today | Apache-2.0 | ||
| 18 | GenkitOpen-source framework for AI features and agents with flows, tool calling, RAG and a local developer UI; works with Gemini and other providers. | Framework | 6.6k | today | Apache-2.0 | |||
| 19 | strataMCP server for progressive tool usage at any scale (see https://klavis.ai) | MCP server | klavis.ai | 5.8k | 16d ago | Apache-2.0 | ||
| 20 | AG2Community fork of AutoGen 0.2 for multi-agent conversations, group chats and tool use, maintained independently of Microsoft. | Framework | ag2ai | 5.1k | today | Apache-2.0 | ||
| 21 | aciACI.dev is the open source tool-calling platform that hooks up 600+ tools into any agentic IDE or custom AI agent through direct function calling or a unified MCP server. The birthplace of VibeOps. | Agent project | aipotheosis-labs | 4.9k | 20d ago | Apache-2.0 | ||
| 22 | ai-agents-from-scratchDemystify AI agents by building them yourself. Local LLMs, no black boxes, real understanding of function calling, memory, and ReAct patterns. | Agent project | pguso | 4.9k | 4mo ago | MIT | ||
| 23 | archestraEnterprise AI Platform with guardrails, MCP registry, gateway & orchestrator | MCP server | archestra-ai | 4.5k | today | — | ||
| 24 | arcboxRun AI agents on real and isolated machines — own kernel, filesystem, and network — with <100ms boot. Local first, OCI compatible, pure Rust. | Agent project | arcboxlabs | 4.3k | 5d ago | Apache-2.0 | ||
| 25 | langroidHarness LLMs with Multi-Agent Programming | Framework | langroid | 4.2k | today | MIT | ||
| 26 | Claude APIAPI for Claude models with tool use, extended thinking, prompt caching, a remote MCP connector and server-side tools such as web search and code execution. | API | Anthropic | 4.1k | 1d ago | Apache-2.0 | ||
| 27 | Gemini APIDeveloper API for Gemini models with long context, multimodal input, function calling and grounding with Google Search. | API | 4.1k | today | Apache-2.0 | |||
| 28 | yu-ai-agent编程导航 AI 开发实战新项目,基于 Spring Boot 3 + Java 21 + Spring AI 构建 AI 恋爱大师应用和 ReAct 模式自主规划智能体YuManus,覆盖 AI 大模型接入、Spring AI 核心特性、Prompt 工程和优化、RAG 检索增强、向量数据库、Tool Calling 工具调用、MCP 模型上下文协议、AI Agent 开发(Manas Java 实现)、Cursor AI 工具等核心知识。用一套教程将程序员必知必会的 AI 技术一网打尽,帮你成为 AI 时代企业的香饽饽,给你的简历和求职大幅增加竞争力。 | Infrastructure | liyupi | 2.8k | 10mo ago | — | ||
| 29 | little-coderA harness optimized to smaller LLMs | Agent project | itayinbarr | 2.7k | 2mo ago | Apache-2.0 | ||
| 30 | VCPToolBoxVCP 部署在 AI 模型 API 与前端应用之间,是面向AGI OS开发和探索的工业级基建示范项目。通过统一指令协议、多层级持久化记忆、分布式插件引擎及多 Agent 协作框架,将原本“无状态、无记忆、无工具调用能力”的大语言模型,彻底改造成拥有永久自我意识、物理世界操作权及群体协作智能的完整智能体系统。 | Framework | lioensky | 2.5k | 1d ago | — | ||
| 31 | solidworks-automation-skillReliable AI Skill + MCP toolkit for agent-driven desktop CAD automation. | MCP server | wzyn20051216 | 2.2k | today | MIT | ||
| 32 | yomo🦖 Serverless AI Agent Framework with Geo-distributed Edge AI Infra. | Framework | yomorun | 1.9k | 1d ago | — | ||
| 33 | DiveDive is an open-source MCP Host Desktop Application that seamlessly integrates with any LLMs supporting function calling capabilities. ✨ | MCP server | OpenAgentPlatform | 1.8k | 3mo ago | MIT | ||
| 34 | tooluniverse2,700+ scientific tools for AI scientists: life science, research, literature, and more. | MCP server | mims-harvard | 1.8k | 3d ago | Apache-2.0 | ||
| 35 | pi-from-scratch600 行 TypeScript 写成的超级迷你版 pi,让你轻松从 0 写出属于你的 pi-agent | Agent project | SaladDay | 1.3k | 28d ago | MIT | ||
| 36 | agentdockSecure MCP runtime for AI agents to operate local machines, servers, and containers with multi-device orchestration. | MCP server | uvwt | 1.3k | 3d ago | Apache-2.0 | ||
| 37 | search2aiSelf-hosted search gateway for AI agents. Unified search, news and page extraction with provider fallback. REST API, MCP server and TypeScript library. | MCP server | fatwang2 | 1.3k | 4d ago | MIT | ||
| 38 | LightAgentLightAgent: Lightweight Python framework for OpenAI-compatible agents with tools, memory, guardrails, tracing, lifecycle hooks, multi-agent collaboration, and workflows. | Framework | wanxingai | 1.2k | 6d ago | Apache-2.0 | ||
| 39 | awesome-ai-sdksA database of SDKs, frameworks, libraries, and tools for creating, monitoring, debugging and deploying autonomous AI agents | Infrastructure | e2b-dev | 1.2k | 4mo ago | — | ||
| 40 | enterprise-deep-researchSalesforce Enterprise Deep Research | Framework | SalesforceAIResearch | 1.2k | 25d ago | Apache-2.0 |
Most common integrations
Third-party services named by products in this node, whole catalogue.
| Integration | Products | Share |
|---|---|---|
| AWS | 2 | 0.7% |
| Blender | 2 | 0.7% |
| GitHub | 2 | 0.7% |
| OpenAPI | 2 | 0.7% |
| PostgreSQL | 2 | 0.7% |
| Apache Kafka | 1 | 0.4% |
| Discord | 1 | 0.4% |
| Figma | 1 | 0.4% |
| Firebase | 1 | 0.4% |
| MongoDB | 1 | 0.4% |
| MySQL | 1 | 0.4% |
| Salesforce | 1 | 0.4% |
| Solana | 1 | 0.4% |
| Telegram | 1 | 0.4% |
| Vercel | 1 | 0.4% |
Launches per month
New Tool Use & Function Calling products per monthSRC
Whole catalogue, by publication month, 2024-11 to 2026-11. The last month is partial; undated products are excluded.
Show the data as a table
| Products published (products) | |
|---|---|
| 2024-11 | 1 |
| 2025-01 | 3 |
| 2025-03 | 6 |
| 2025-04 | 4 |
| 2025-05 | 2 |
| 2025-07 | 5 |
| 2025-08 | 1 |
| 2025-09 | 3 |
| 2025-10 | 3 |
| 2025-11 | 4 |
| 2025-12 | 3 |
| 2026-01 | 3 |
| 2026-02 | 22 |
| 2026-03 | 8 |
| 2026-04 | 14 |
| 2026-05 | 12 |
| 2026-06 | 17 |
| 2026-07 | 17 |
| 2026-08 | 19 |
| 2026-09 | 32 |
| 2026-10 | 28 |
| 2026-11 | 28 |
Related opportunities
Opportunity radarScored gaps for this node or anything beneath it. Scores rank hypotheses; they do not predict demand.
| Opportunity | Competition | Confidence | Score |
|---|---|---|---|
| Coverage gap: Code Execution SandboxesCode Execution Sandboxes | High | Low | 54.1 |
| Coverage gap: Structured OutputsStructured Outputs | High | Low | 52.1 |
Limits of these numbers
- Products are placed here by whole-word keyword rules on their name, description and repository topics, and inherit every ancestor of a matched node. Products that do this without naming it are missed; ambiguous words cause false matches. Methodology
- About half of the catalogue matches no node at all, so absolute counts understate every category.
- Launch counts use the publication date each source reports; for auto-discovered repositories that is the repository’s creation date.
- Stars and concentration cover only products with their own repository (not a subfolder of a larger one). They measure attention, not adoption.
Page data last updated .