# AgentToolDrop

> A curated, ranked directory of 71 tools for building with AI agents: coding agents, MCP servers, agent frameworks, browser agents, memory, evals, sandboxes and model APIs. Every entry says what the tool does, when to use it, when to skip it, and how to install it (snippets copied from the project's own docs), with GitHub signals (stars, last push, latest release) and the date they were fetched.

Generated 2026-10-15T03:12:48.485Z. GitHub signals last fetched 2026-10-15T01:19:06.303Z.

## How to query

- Full catalog as JSON (all fields, live signals with fetch dates and status, drop score): https://agenttooldrop.com/tools.json
- One tool as markdown (frontmatter, effect, why, use when, skip if, install, signals): https://agenttooldrop.com/tools/{slug}.md
- Every tool in one markdown file: https://agenttooldrop.com/llms-full.txt
- Human pages: https://agenttooldrop.com/tools/{slug} and https://agenttooldrop.com/category/{category}
- To pick a tool: filter tools.json by `category` and `tags`, read `use_when` and `skip_if`, then copy `install.snippet` (verified at `install.source_url`).

## Categories

- [Coding agents](https://agenttooldrop.com/category/coding-agent): `coding-agent`, 11 tools. Agents that read a codebase, edit files and run commands, in the terminal, the editor or the cloud.
- [MCP servers](https://agenttooldrop.com/category/mcp-server): `mcp-server`, 13 tools. Model Context Protocol servers that give an agent tools and data: browsers, repos, docs, search, payments.
- [Agent frameworks](https://agenttooldrop.com/category/agent-framework): `agent-framework`, 14 tools. Libraries and SDKs for writing your own tool-using agents and multi-step workflows.
- [Browser and computer use](https://agenttooldrop.com/category/browser-agent): `browser-agent`, 4 tools. Agents and infrastructure that drive a real browser or desktop on your behalf.
- [Memory](https://agenttooldrop.com/category/memory): `memory`, 4 tools. Long-term memory layers and knowledge graphs that let agents remember across sessions.
- [Evals and observability](https://agenttooldrop.com/category/eval-observability): `eval-observability`, 7 tools. Tracing, evaluation and red-teaming so you can see what an agent did and whether it was right.
- [Sandboxes](https://agenttooldrop.com/category/sandbox): `sandbox`, 3 tools. Isolated environments where agents can run generated code without touching your machine.
- [Models and APIs](https://agenttooldrop.com/category/model-api): `model-api`, 9 tools. Model APIs, routers, local inference servers and libraries for calling LLMs from code.
- [Devtools](https://agenttooldrop.com/category/devtool): `devtool`, 6 tools. Utilities agentic engineers use around their agents: MCP tooling, context packing, specs, usage tracking.

## Tools (ranked by drop score)

- [Claude Code](https://agenttooldrop.com/tools/claude-code.md) `claude-code` (coding-agent): Reads your codebase, edits files, runs commands and handles git work from the terminal, IDE, desktop or web. https://claude.com/product/claude-code
- [OpenAI Codex CLI](https://agenttooldrop.com/tools/openai-codex-cli.md) `openai-codex-cli` (coding-agent): Runs OpenAI's coding agent locally to inspect your repo, edit files and run commands from the terminal. https://learn.chatgpt.com/docs/codex/cli
- [Context7](https://agenttooldrop.com/tools/context7.md) `context7` (mcp-server): Pulls current, version-specific library docs and code examples into your agent's context. https://context7.com
- [LangGraph](https://agenttooldrop.com/tools/langgraph.md) `langgraph` (agent-framework): Orchestrates long-running, stateful agents as graphs with durable execution and human-in-the-loop. https://www.langchain.com/langgraph
- [Langfuse](https://agenttooldrop.com/tools/langfuse.md) `langfuse` (eval-observability): Traces LLM calls and agent steps, then runs evals, datasets and prompt versioning on top of them. https://langfuse.com
- [Ollama](https://agenttooldrop.com/tools/ollama.md) `ollama` (model-api): Runs open models locally behind a CLI and a REST API, with an OpenAI-compatible endpoint. https://ollama.com
- [OpenCode](https://agenttooldrop.com/tools/opencode.md) `opencode` (coding-agent): Runs an open-source coding agent in your terminal, desktop app or browser with the LLM provider you choose. https://opencode.ai/
- [Spec Kit](https://agenttooldrop.com/tools/spec-kit.md) `spec-kit` (devtool): Gives coding agents structured processes for spec-driven builds, bug fixes and idea assessment. https://github.github.com/spec-kit/
- [LangChain](https://agenttooldrop.com/tools/langchain.md) `langchain` (agent-framework): Provides one interface and a large integration library for models, embeddings, vector stores and tools. https://docs.langchain.com/oss/python/langchain/overview
- [OpenHands](https://agenttooldrop.com/tools/openhands.md) `openhands` (coding-agent): Runs OpenHands, Claude Code, Codex and other ACP agents from a self-hosted control center. https://www.openhands.dev/
- [Cline](https://agenttooldrop.com/tools/cline.md) `cline` (coding-agent): Runs an approval-based coding agent in VS Code, JetBrains, the terminal or a desktop app with any model. https://cline.bot/
- [LiteLLM](https://agenttooldrop.com/tools/litellm.md) `litellm` (model-api): Calls 100+ LLM providers in the OpenAI format, as a Python SDK or a self-hosted AI gateway. https://docs.litellm.ai
- [Anthropic Agent Skills](https://agenttooldrop.com/tools/anthropic-skills.md) `anthropic-skills` (devtool): Provides Anthropic's example Agent Skills, the document skills behind Claude, and the skills spec. https://github.com/anthropics/skills
- [Browser Use](https://agenttooldrop.com/tools/browser-use.md) `browser-use` (browser-agent): Lets an LLM drive a real browser to complete multi-step web tasks from Python code or a CLI. https://browser-use.com
- [Mastra](https://agenttooldrop.com/tools/mastra.md) `mastra` (agent-framework): Builds TypeScript agents and graph-based workflows with memory, RAG, evals and model routing. https://mastra.ai/
- [Playwright MCP](https://agenttooldrop.com/tools/playwright-mcp.md) `playwright-mcp` (mcp-server): Lets an agent drive a real browser with Playwright using accessibility snapshots, not screenshots. https://github.com/microsoft/playwright-mcp
- [Vercel AI SDK](https://agenttooldrop.com/tools/vercel-ai-sdk.md) `vercel-ai-sdk` (agent-framework): Gives TypeScript apps one API for text, structured output, tool-calling agents and chat UI hooks. https://ai-sdk.dev/
- [Gemini CLI](https://agenttooldrop.com/tools/gemini-cli.md) `gemini-cli` (coding-agent): Brings Gemini models into the terminal to query and edit codebases, run shell commands and search the web. https://geminicli.com/
- [Mem0](https://agenttooldrop.com/tools/mem0.md) `mem0` (memory): Adds long-term user, session and agent memory to AI apps through a Python or TypeScript SDK. https://mem0.ai
- [GitHub MCP Server](https://agenttooldrop.com/tools/github-mcp-server.md) `github-mcp-server` (mcp-server): Connects agents to GitHub repos, issues, pull requests, Actions runs and security alerts. https://github.com/github/github-mcp-server
- [Pydantic AI](https://agenttooldrop.com/tools/pydantic-ai.md) `pydantic-ai` (agent-framework): Runs a typed Python agent loop with validated tools and outputs, and models swapped by a string. https://pydantic.dev/docs/ai/overview/
- [Chrome DevTools MCP](https://agenttooldrop.com/tools/chrome-devtools-mcp.md) `chrome-devtools-mcp` (mcp-server): Lets a coding agent control and inspect a live Chrome browser through DevTools. https://github.com/ChromeDevTools/chrome-devtools-mcp
- [E2B](https://agenttooldrop.com/tools/e2b.md) `e2b` (sandbox): Runs AI-generated code in isolated cloud sandboxes you start and control from Python or JavaScript. https://e2b.dev
- [Arize Phoenix](https://agenttooldrop.com/tools/arize-phoenix.md) `arize-phoenix` (eval-observability): Traces LLM and agent apps over OpenTelemetry and runs evals, datasets and experiments in a local UI. https://arize.com/phoenix/
- [CrewAI](https://agenttooldrop.com/tools/crewai.md) `crewai` (agent-framework): Orchestrates teams of role-based agents (Crews) and event-driven workflows (Flows) in Python. https://crewai.com/
- [FastMCP](https://agenttooldrop.com/tools/fastmcp.md) `fastmcp` (devtool): Builds MCP servers, clients and interactive apps in Python from plain decorated functions. https://gofastmcp.com
- [Model Context Protocol](https://agenttooldrop.com/tools/model-context-protocol.md) `model-context-protocol` (mcp-server): Defines an open standard for connecting AI applications to tools, data sources and workflows. https://modelcontextprotocol.io
- [OpenAI Agents SDK](https://agenttooldrop.com/tools/openai-agents-sdk.md) `openai-agents-sdk` (agent-framework): Builds multi-agent workflows with handoffs, guardrails, sessions and built-in tracing. https://openai.github.io/openai-agents-python/
- [Promptfoo](https://agenttooldrop.com/tools/promptfoo.md) `promptfoo` (eval-observability): Runs evals and red-team scans against prompts, models and agents from a CLI, with a web results viewer. https://www.promptfoo.dev
- [Claude Agent SDK](https://agenttooldrop.com/tools/claude-agent-sdk.md) `claude-agent-sdk` (agent-framework): Embeds Claude Code's agent loop, built-in tools and context management in Python or TypeScript apps. https://code.claude.com/docs/en/agent-sdk/overview
- [llama.cpp](https://agenttooldrop.com/tools/llama-cpp.md) `llama-cpp` (model-api): Runs LLM and VLM inference locally in plain C/C++, with a CLI and an OpenAI-compatible server. https://llama.app
- [Fetch MCP Server](https://agenttooldrop.com/tools/fetch-mcp-server.md) `fetch-mcp-server` (mcp-server): Fetches a URL and returns the page as markdown in chunks the model can page through. https://github.com/modelcontextprotocol/servers/tree/main/src/fetch
- [Filesystem MCP Server](https://agenttooldrop.com/tools/filesystem-mcp-server.md) `filesystem-mcp-server` (mcp-server): Lets an agent read, write, edit, move and search files inside directories you allow. https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem
- [MCP Inspector](https://agenttooldrop.com/tools/mcp-inspector.md) `mcp-inspector` (devtool): Tests and debugs MCP servers from a web UI, a scriptable CLI or a terminal UI. https://modelcontextprotocol.io/docs/tools/inspector
- [vLLM](https://agenttooldrop.com/tools/vllm.md) `vllm` (model-api): Serves open-weight models at high throughput behind an OpenAI-compatible API server. https://vllm.ai
- [Claude API](https://agenttooldrop.com/tools/anthropic-api.md) `anthropic-api` (model-api): Gives programmatic access to Claude models and Claude Managed Agents over a REST API with official SDKs. https://claude.com/platform/api
- [Agno](https://agenttooldrop.com/tools/agno.md) `agno` (agent-framework): Builds Python agents, teams and workflows and serves them as an API with the AgentOS runtime. https://docs.agno.com/
- [Graphiti](https://agenttooldrop.com/tools/graphiti.md) `graphiti` (memory): Builds temporal knowledge graphs from agent conversations and data, tracking how facts change over time. https://help.getzep.com/graphiti
- [Supabase MCP Server](https://agenttooldrop.com/tools/supabase-mcp.md) `supabase-mcp` (mcp-server): Lets an AI assistant work with Supabase projects: tables, migrations, SQL, logs and Edge Functions. https://supabase.com/mcp
- [Cursor](https://agenttooldrop.com/tools/cursor.md) `cursor` (coding-agent): Gives you a code editor and CLI with a built-in agent that plans, edits and reviews changes across your repo. https://cursor.com/
- [GitHub Copilot](https://agenttooldrop.com/tools/github-copilot.md) `github-copilot` (coding-agent): Suggests code, answers questions and works on assigned tasks across your IDE, terminal and GitHub. https://github.com/features/copilot
- [OpenRouter](https://agenttooldrop.com/tools/openrouter.md) `openrouter` (model-api): Routes requests to hundreds of models across providers through one OpenAI-compatible API endpoint. https://openrouter.ai
- [goose](https://agenttooldrop.com/tools/goose.md) `goose` (coding-agent): Runs a general-purpose AI agent on your machine as a desktop app, CLI or embeddable API. https://goose-docs.ai/
- [DSPy](https://agenttooldrop.com/tools/dspy.md) `dspy` (model-api): Builds LM programs from typed signatures and modules, then optimizes their prompts against a metric. https://dspy.ai
- [Cognee](https://agenttooldrop.com/tools/cognee.md) `cognee` (memory): Turns documents, code and conversations into a knowledge graph that agents store to and recall from. https://www.cognee.ai
- [Opik](https://agenttooldrop.com/tools/opik.md) `opik` (eval-observability): Traces agent runs and LLM calls, scores them with LLM-as-a-judge metrics and monitors production. https://www.comet.com/docs/opik/
- [Google Agent Development Kit (ADK)](https://agenttooldrop.com/tools/google-adk.md) `google-adk` (agent-framework): Builds, evaluates and deploys Python agents and graph-based workflows, optimized for Gemini. https://adk.dev/
- [ccusage](https://agenttooldrop.com/tools/ccusage.md) `ccusage` (devtool): Reports token usage and costs for Claude Code, Codex and other coding agent CLIs from local data. https://ccusage.com
- [Crush](https://agenttooldrop.com/tools/crush.md) `crush` (coding-agent): Runs a terminal coding agent from Charm that works with your choice of LLM and uses LSPs for context. https://github.com/charmbracelet/crush
- [LlamaIndex](https://agenttooldrop.com/tools/llamaindex.md) `llamaindex` (agent-framework): Connects LLM agents to your data with connectors, indexes, query engines and event-driven workflows. https://developers.llamaindex.ai/python/framework/
- [UI-TARS Desktop](https://agenttooldrop.com/tools/ui-tars-desktop.md) `ui-tars-desktop` (browser-agent): Controls your computer and browser from screenshots, with mouse and keyboard actions chosen by a vision model. https://github.com/bytedance/UI-TARS-desktop
- [Strands Agents](https://agenttooldrop.com/tools/strands-agents.md) `strands-agents` (agent-framework): Runs a model-driven agent loop in your own process, in Python or TypeScript, with any model provider. https://strandsagents.com/
- [Firecrawl MCP Server](https://agenttooldrop.com/tools/firecrawl-mcp.md) `firecrawl-mcp` (mcp-server): Lets agents search, scrape, crawl and interact with web pages through Firecrawl, returning markdown or JSON. https://docs.firecrawl.dev/mcp-server
- [Letta](https://agenttooldrop.com/tools/letta.md) `letta` (memory): Runs stateful agents that keep memory and identity across sessions and rewrite their own memory over time. https://www.letta.com
- [Inspect](https://agenttooldrop.com/tools/inspect-ai.md) `inspect-ai` (eval-observability): Defines and runs model and agent evaluations in Python from datasets, solvers and scorers. https://inspect.aisi.org.uk/
- [Stripe MCP and Agent Toolkit](https://agenttooldrop.com/tools/stripe-agent-toolkit.md) `stripe-agent-toolkit` (mcp-server): Lets agents call the Stripe API and search Stripe docs through a hosted MCP server or framework toolkits. https://docs.stripe.com/agents
- [Daytona](https://agenttooldrop.com/tools/daytona.md) `daytona` (sandbox): Runs AI-generated code in hosted sandboxes with their own kernel, filesystem and network stack. https://www.daytona.io
- [Stagehand](https://agenttooldrop.com/tools/stagehand.md) `stagehand` (browser-agent): Adds natural-language act, observe and extract calls to Playwright-style browser code in TS, Python and Go. https://www.stagehand.dev
- [LangSmith](https://agenttooldrop.com/tools/langsmith.md) `langsmith` (eval-observability): Traces, evaluates and monitors LLM apps and agents on LangChain's hosted platform via Python and JS SDKs. https://www.langchain.com/langsmith/observability
- [Repomix](https://agenttooldrop.com/tools/repomix.md) `repomix` (devtool): Packs a repository into one AI-friendly file in XML, Markdown, JSON or plain text. https://repomix.com
- [LLM](https://agenttooldrop.com/tools/llm-cli.md) `llm-cli` (model-api): Runs prompts against remote and local LLMs from the command line or Python, logging calls to SQLite. https://llm.datasette.io
- [DeepEval](https://agenttooldrop.com/tools/deepeval.md) `deepeval` (eval-observability): Unit tests LLM apps and agents in a Pytest-style workflow with LLM-as-a-judge and other metrics. https://deepeval.com
- [Framelink MCP for Figma](https://agenttooldrop.com/tools/figma-context-mcp.md) `figma-context-mcp` (mcp-server): Gives a coding agent the layout and styling data behind a Figma file, frame or group link. https://www.framelink.ai/
- [Instructor](https://agenttooldrop.com/tools/instructor.md) `instructor` (model-api): Returns validated Pydantic objects from LLM calls, retrying automatically when validation fails. https://python.useinstructor.com/
- [smolagents](https://agenttooldrop.com/tools/smolagents.md) `smolagents` (agent-framework): Runs agents that write their actions as Python code, with core agent logic in about 1,000 lines. https://huggingface.co/docs/smolagents/index
- [Sentry MCP Server](https://agenttooldrop.com/tools/sentry-mcp.md) `sentry-mcp` (mcp-server): Gives coding agents access to Sentry errors, issues, traces and performance data over MCP. https://mcp.sentry.dev
- [Aider](https://agenttooldrop.com/tools/aider.md) `aider` (coding-agent): Pair programs with an LLM in your terminal, editing files in your git repo and committing each change. https://aider.chat/
- [Exa MCP Server](https://agenttooldrop.com/tools/exa-mcp.md) `exa-mcp` (mcp-server): Gives agents Exa web search and page fetching as MCP tools, with optional multi-step research runs. https://exa.ai/docs/get-started/exa-mcp
- [Steel Browser](https://agenttooldrop.com/tools/steel-browser.md) `steel-browser` (browser-agent): Runs a self-hostable browser API that manages Chrome sessions agents connect to with Puppeteer or Playwright. https://steel.dev
- [AutoGen](https://agenttooldrop.com/tools/autogen.md) `autogen` (agent-framework): Builds multi-agent AI apps that act autonomously or alongside humans; now in maintenance mode. https://microsoft.github.io/autogen/
- [Container Use](https://agenttooldrop.com/tools/container-use.md) `container-use` (sandbox): Gives each coding agent its own container and git branch so several can work in parallel without conflicts. https://container-use.com