Agent Infrastructure
The frameworks, runtimes, models, security and observability layers agents run on. As of 2026-11-21, 8,733 indexed products match this ecosystem (2,364 with full profiles); 4,234 were published in the last 90 days, against 1,923 in the 90 days before, slower than the catalogue as a whole. By GitHub stars, MarkItDown leads, holding 3% of the stars of the 5,705 products here with their own repository. The most common form is MCP servers (83% of indexed products).
EcosystemSegment label: Crowded, slowing DER
Segment statistics
Whole catalogue (all 8,733 indexed products, not only profiled ones). Calculated 2026-11-21.
Crowded, slowing. Among the largest fifth of nodes at its level, but its launches grew more slowly than the catalogue’s. Size is ranked against the other ecosystem 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 0.007, attention is spread as if across about 142 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 | 7,275 | 83.3% | 1,044 |
| Agent skills | 333 | 3.8% | 329 |
| Infrastructure | 288 | 3.3% | 282 |
| Frameworks | 242 | 2.8% | 235 |
| Plugins | 199 | 2.3% | 156 |
| Libraries | 107 | 1.2% | 96 |
| Rules packs | 95 | 1.1% | 31 |
| Agent projects | 93 | 1.1% | 89 |
| IDE extensions | 28 | 0.32% | 29 |
| SaaS products | 18 | 0.21% | 18 |
| Developer tools | 15 | 0.17% | 15 |
| AI APIs | 13 | 0.15% | 13 |
| Agent SDKs | 9 | 0.10% | 9 |
| Protocols & standards | 7 | 0.08% | 7 |
| Platforms | 7 | 0.08% | 7 |
| Automation platforms | 4 | 0.05% | 4 |
6 categories
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.
| Category | Indexed | Profiled | New, 90 days | Label |
|---|---|---|---|---|
| Agent CommerceHow agents pay, bill, meter usage and sell through marketplaces. | 3,621 | 199 | 2,284 | Crowded |
| Agent SecurityIdentity, access, governance and threat protection for agents. | 1,481 | 469 | 650 | Lagging the catalogue |
| Agent Runtime & DeploymentHosting agents in production and keeping their state. | 1,461 | 487 | 614 | Lagging the catalogue |
| Agent Frameworks & OrchestrationLibraries and platforms for building and coordinating agents. | 1,192 | 735 | 311 | Lagging the catalogue |
| Models & GatewaysAccessing, routing, serving and tuning language models. | 1,115 | 601 | 343 | Lagging the catalogue |
| Agent ObservabilitySeeing, measuring and evaluating what agents do. | 951 | 440 | 367 | Lagging the catalogue |
Top 40 products by GitHub stars
Search the directoryProfiled products only (2,364 of 8,733), 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 | MarkItDownPython utility that converts PDF, Office, HTML, images and audio files into Markdown for LLM consumption. | Library | microsoft | 197k | 14d ago | MIT | ||
| 2 | ponytailMakes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. | Agent skill | DietrichGebert | 196k | 23d ago | MIT | ||
| 3 | DifyPlatform for building LLM apps and agent workflows visually with RAG pipelines, model management and app publishing. | Automation platform | LangGenius | 162k | today | LicenseRef-Dify-Open-Source-License | ||
| 4 | LangflowVisual builder for agent and RAG flows using drag-and-drop components, with flows deployable as APIs or MCP servers. | Automation platform | DataStax | 157k | today | MIT | ||
| 5 | LangChainFramework of model, tool, retriever and agent abstractions with hundreds of provider integrations; its agent API now runs on LangGraph. | Framework | LangChain | 150k | today | MIT | ||
| 6 | claude-memPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More | Agent skill | thedotmack | 119k | today | Apache-2.0 | ||
| 7 | labelhead-artist-momentumTrending hip-hop artist momentum scores across four cultural dimensions. | MCP server | IncorporatedPartners | 116k | today | MIT | ||
| 8 | RAGFlowSelf-hostable RAG engine with deep document parsing, template-based chunking, citations and an agent builder. | Framework | infiniflow | 94k | 1d ago | Apache-2.0 | ||
| 9 | OpenHandsOpen-source platform for AI coding agents comprising a Python Software Agent SDK and Agent Server, the Agent Canvas browser client, and a managed OpenHands Cloud with GitHub, GitLab, Slack and Jira integrations. | Platform | OpenHands (All Hands AI) | 94k | today | MIT | ||
| 10 | OmniRouteNever stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors | Agent project | cheaper-inference | 89k | today | MIT | ||
| 11 | MiroFishA Simple and Universal Swarm Intelligence Engine, Predicting Anything. 简洁通用的群体智能引擎,预测万物 | Infrastructure | 666ghj | 86k | 1mo ago | AGPL-3.0 | ||
| 12 | dsh-our-free-model在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费,不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, DeepSeek V4.1 Flash and Kimi K3 among them. Completely free, with no usage cap. | Agent project | Ebony-Vinyl | 83k | today | — | ||
| 13 | claude-flowAI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev. | MCP server | ruvnet | 82k | today | MIT | ||
| 14 | ai-agents-for-beginners18 Lessons to Get Started Building AI Agents | Agent project | microsoft | 81k | 4d ago | MIT | ||
| 15 | Prompt-Engineering-Guide🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents. | Agent project | dair-ai | 80k | 8mo ago | MIT | ||
| 16 | system_prompts_leaksLeaked system prompts of ChatGPT, Claude, Gemini and more. The hidden system prompt of every popular chatbot, in full, every version since May 2025. Also Anthropic Claude Code, OpenAI Codex, Google Antigravity, xAI Grok, Meta Muse, Moonshot Kimi, Alibaba Qwen, Z.ai GLM ZCode, MiniMax, Cursor, Perplexity, DeepSeek and 30+ others. Updated daily. | Agent project | asgeirtj | 72k | today | CC0-1.0 | ||
| 17 | humanizerAgent skill that removes signs of AI-generated writing from text | Agent skill | blader | 72k | 2d ago | MIT | ||
| 18 | DaytonaSandbox infrastructure that starts isolated environments for running AI-generated code quickly, with SDKs for files, git and processes. | Infrastructure | Daytona | 71k | 4mo ago | — | ||
| 19 | Mem0Memory layer that extracts, stores and retrieves long-term facts about users and sessions for agents, available as an open-source library or hosted platform. | Library | Mem0 | 69k | 3d ago | Apache-2.0 | ||
| 20 | anything-llmStop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience | Agent project | Mintplex-Labs | 68k | today | MIT | ||
| 21 | LiteLLMPython SDK and proxy server that expose 100+ model providers through an OpenAI-compatible API with keys, budgets, rate limits, fallbacks and logging. | Infrastructure | BerriAI | 64k | today | MIT | ||
| 22 | AutoGenFramework for multi-agent conversations and event-driven agent systems that popularised conversable-agent patterns. | Framework | microsoft | 62k | 9d ago | CC-BY-4.0 | ||
| 23 | CrewAIPython framework for role-based agent teams (crews) and event-driven Flows, independent of LangChain; paired with a commercial platform for deployment. | Framework | CrewAI | 61k | today | MIT | ||
| 24 | ai-agent-book《深入理解 AI Agent:设计原理与工程实践》(李博杰 著)开源主仓库:全书正文、编译版 PDF 与按章配套代码 | Agent project | bojieli | 59k | 1mo ago | Apache-2.0 | ||
| 25 | gooseOpen-source, local general-purpose AI agent (desktop app and CLI) for coding, research, writing and automation that works with any LLM and is extended through MCP extensions, YAML recipes and subagents. | Platform | Agentic AI Foundation (originally Block) | 57k | 1d ago | Apache-2.0 | ||
| 26 | cherry-studioYour AI workspace for models, agents, and productivity. | Agent skill | CherryHQ | 54k | today | AGPL-3.0 | ||
| 27 | LlamaIndexFramework for building agents and workflows over your data, with connectors, indexing, retrieval and an event-driven Workflows engine. | Framework | LlamaIndex | 53k | 1d ago | MIT | ||
| 28 | herdrthe runtime your coding agents live on | Agent project | herdrdev | 52k | 1d ago | Apache-2.0 | ||
| 29 | nanobotYour lightweight open-source personal agent. | MCP server | HKUDS | 50k | today | MIT | ||
| 30 | CowAgentOpen-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install. | Agent project | zhayujie | 48k | today | MIT | ||
| 31 | iFixAiIndependent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is supposed to do? With iFixAi you can have this answer in less than 120 seconds. | Infrastructure | ifixai-ai | 47k | 1d ago | Apache-2.0 | ||
| 32 | LangGraphLow-level orchestration framework that models agents as stateful graphs with checkpointing, human-in-the-loop interrupts and durable execution. | Framework | LangChain | 45k | today | MIT | ||
| 33 | Kong AI GatewaySet of AI plugins for the Kong API gateway that proxy and govern LLM traffic with prompt guards, semantic caching, routing and MCP traffic controls. | Infrastructure | Kong | 44k | 1d ago | Apache-2.0 | ||
| 34 | AgnoPython framework and runtime for agents and multi-agent teams with memory, knowledge and a production API server (AgentOS). | Framework | Agno | 43k | 1d ago | Apache-2.0 | ||
| 35 | openhumanThe fastest, cheapest, most efficient open-source agent harness. Run more than 500 agents on a $10 VPS. | Agent project | tinyhumansai | 43k | 1d ago | GPL-3.0 | ||
| 36 | agentsPlugins, agents, skills, and commands for Claude Code, Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, and Pi, written once in Markdown and adapted to each tool by generators with validation and evals. | Agent skill | wshobson | 43k | 3d ago | MIT | ||
| 37 | BettaFish微舆:人人可用的多Agent舆情分析助手,打破信息茧房,还原舆情原貌,预测未来走向,辅助决策!从0实现,不依赖任何框架。 | Framework | 666ghj | 43k | 1mo ago | GPL-2.0 | ||
| 38 | OpenVikingSelf-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills. | Infrastructure | volcengine | 43k | today | AGPL-3.0 | ||
| 39 | awesome-copilotCommunity-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot. | Agent skill | github | 41k | 1d ago | MIT | ||
| 40 | CodewhaleAn open-source coding agent for your terminal, written in Rust. It works with any model, can split one task across several agents, and every line of it is yours to read. | Agent project | codewhale-hq | 41k | today | MIT |
Most common integrations
Third-party services named by products in this node, whole catalogue.
| Integration | Products | Share |
|---|---|---|
| Solana | 181 | 2.1% |
| GitHub | 71 | 0.8% |
| Stripe | 40 | 0.5% |
| PostgreSQL | 39 | 0.4% |
| YouTube | 38 | 0.4% |
| Ethereum | 32 | 0.4% |
| AWS | 30 | 0.3% |
| 27 | 0.3% | |
| Google Cloud | 24 | 0.3% |
| Kubernetes | 24 | 0.3% |
| Obsidian | 24 | 0.3% |
| Azure | 23 | 0.3% |
| Telegram | 21 | 0.2% |
| Android | 17 | 0.2% |
| Shopify | 17 | 0.2% |
Launches per month
New Agent Infrastructure 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 | 8 |
| 2024-12 | 12 |
| 2025-01 | 18 |
| 2025-02 | 15 |
| 2025-03 | 29 |
| 2025-04 | 34 |
| 2025-05 | 37 |
| 2025-06 | 20 |
| 2025-07 | 30 |
| 2025-08 | 22 |
| 2025-09 | 48 |
| 2025-10 | 51 |
| 2025-11 | 35 |
| 2025-12 | 48 |
| 2026-01 | 82 |
| 2026-02 | 256 |
| 2026-03 | 375 |
| 2026-04 | 546 |
| 2026-05 | 425 |
| 2026-06 | 471 |
| 2026-07 | 805 |
| 2026-08 | 814 |
| 2026-09 | 935 |
| 2026-10 | 1,750 |
| 2026-11 | 1,299 |
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: Task PlanningTask Planning | Medium | Low | 63.0 |
| Coverage gap: User Preference MemoryUser Preference Memory | Medium | Low | 61.0 |
| Coverage gap: Session PersistenceSession Persistence | High | Low | 60.0 |
| Coverage gap: TypeScript Agent FrameworksTypeScript Agent Frameworks | High | Low | 59.6 |
| Coverage gap: No-code Agent BuildersNo-code Agent Builders | High | Low | 59.2 |
| Coverage gap: Background AgentsBackground Agents | Medium | Low | 59.1 |
| Coverage gap: LoggingLogging | Medium | Low | 59.1 |
| Coverage gap: Cross-model ConsultationCross-model Consultation | High | Low | 58.8 |
| Coverage gap: Stateful Workflow GraphsStateful Workflow Graphs | High | Low | 58.4 |
| Coverage gap: Red TeamingRed Teaming | High | Low | 58.1 |
| Coverage gap: Agent HostingAgent Hosting | Medium | Low | 57.5 |
| Coverage gap: Model RoutingModel Routing | High | Low | 57.3 |
| Coverage gap: Performance MonitoringPerformance Monitoring | High | Low | 55.5 |
| Coverage gap: Usage MeteringUsage Metering | High | Low | 55.1 |
| Coverage gap: Code Execution SandboxesCode Execution Sandboxes | High | Low | 54.1 |
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 .