# AgentPick

> A directory of AI agent tools ranked by use case. Each tool gets a short verdict and cites its source. Rankings are earned only. Paid placements are labelled "Featured" and never change rank. Tools marked "Tested: no" have provisional verdicts based on public sources.

- [Full directory as text](https://agentpick.space/llms-full.txt)
- [How we test](https://agentpick.space/how-we-test/)
- [Tool updates](https://agentpick.space/updates/)
- [How we rank](https://agentpick.space/how-we-rank/): a fixed rubric (job fit 25%, output quality 25%, ease of start 15%, pricing & value 15%, reliability & control 10%, momentum & openness 10%)

## Categories

### [Vibe / web app coding](https://agentpick.space/category/vibe-coding/)
Prompt-to-app builders.
- Nothing new this week.

### [Coding agents & IDE assistants](https://agentpick.space/category/coding-agents/)
Agents that edit real codebases.
- Listed, not ranked: [Claude Code 2.1.300](https://agentpick.space/tool/claude-code/) (released 2026-10-14, Tested: no): An incremental release aimed at people customizing Claude Code with mods and monitoring it with OpenTelemetry. Nothing changes for everyday use. Source: https://github.com/anthropics/claude-code/releases/tag/v2.1.300
- Listed, not ranked: [Kilo Code 7.8.10](https://agentpick.space/tool/kilo-code/) (released 2026-10-14, Tested: no): A small quality-of-life release. Useful if your agent kept missing generated files you actually wanted it to read. Source: https://github.com/Kilo-Org/kilocode/releases/tag/v7.8.10
- Listed, not ranked: [Vani Code Agent for VS Code](https://agentpick.space/tool/vani-code-agent/) (released 2026-10-14, Tested: no): A careful, approval-first coding agent and an easy pick for VS Code users who want to try another model. It's early-stage: desktop apps and a CLI haven't shipped yet. Source: https://vani.ai/news/vani-code-agent-for-vs-code
- Listed, not ranked: [Cline 4.1.24](https://agentpick.space/tool/cline/) (released 2026-10-13, Tested: no): A maintenance release: mostly model-catalog and default changes. Check your model selection after updating, since defaults changed for many providers. Source: https://github.com/cline/cline/releases/tag/v4.1.24
- Listed, not ranked: [Gemini 4 Argon (incl. GitHub Copilot)](https://agentpick.space/tool/gemini-4-argon/) (released 2026-10-13, Tested: no): It's the new model to try for long refactors and migrations if you already pay for Copilot or AI Ultra. Access on lower Copilot tiers is limited. Source: https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-argon/
- Listed, not ranked: [Gemini CLI 0.64.0](https://agentpick.space/tool/gemini-cli/) (released 2026-10-13, Tested: no): A maintenance release, mostly relevant if you run Gemini CLI as an A2A server or through ACP-based editors. Source: https://github.com/google-gemini/gemini-cli/releases/tag/v0.64.0
- Listed, not ranked: [Muse Wonder / Muse Wonder Full](https://agentpick.space/tool/muse-wonder/) (released 2026-10-12, Tested: no): A strong new contender for heavy, multi-file agent work. The best version sits behind the Power and Maximum Usage plans. Source: https://dev.meta.ai/resources/blog/build-with-muse-wonder
- Listed, not ranked: [OpenHands 1.27.1](https://agentpick.space/tool/openhands/) (released 2026-10-12, Tested: no): A patch release. Update if you run OpenHands in Docker. Source: https://github.com/OpenHands/OpenHands/releases/tag/v1.27.1
- Listed, not ranked: [OpenCode 1.19.0](https://agentpick.space/tool/opencode/) (released 2026-10-11, Tested: no): Mainly a bridge release to OpenCode 2, plus a PTY security check. Run /update when you're ready, since auto-updates are now off. Source: https://github.com/anomalyco/opencode/releases/tag/v1.19.0
- Listed, not ranked: [goose 1.54.0](https://agentpick.space/tool/goose/) (released 2026-10-08, Tested: no): Shareable HTML session exports are handy for reviewing or demoing what an agent did. A solid pick if you want a model-agnostic local agent. Source: https://github.com/aaif-goose/goose/releases/tag/v1.54.0
- Listed, not ranked: [OpenAI Codex CLI 0.162](https://agentpick.space/tool/openai-codex-cli/) (released 2026-10-08, Tested: no): Worktree support is the headline: it makes running parallel agent tasks on one repo cleaner. Worth updating if you run Codex on several tasks at once. Source: https://github.com/openai/codex/releases/tag/rust-v0.162.0

### [Agent building](https://agentpick.space/category/agent-building/)
Frameworks & SDKs.
- Ranked (score 7.7/10): [Pydantic AI 2.56.0](https://agentpick.space/tool/pydantic-ai/) (released 2026-10-13, Tested: yes): The cleanest typed-Python agent framework we've tried. In our offline test, tool calls and structured output worked on the first run with no API key, using the built-in TestModel. Live-model quality isn't tested yet. Source: https://github.com/pydantic/pydantic-ai/releases/tag/v2.56.0
- Ranked (score 7.2/10): [LangChain Deep Agents 0.7.23](https://agentpick.space/tool/deepagents/) (released 2026-10-07, Tested: yes): A small but useful release for anyone building skill-based subagent setups on LangGraph. In our hands-on smoke test it installed cleanly and ran out of the box, and the new skill pinning is in the code. Live-model quality isn't tested yet. Source: https://github.com/langchain-ai/deepagents/releases/tag/deepagents%3D%3D0.7.23
- Listed, not ranked: [CrewAI 1.15.28](https://agentpick.space/tool/crewai/) (released 2026-10-14, Tested: no): A useful debugging release for teams running crews in production: the new eval command targets the hardest question, where a run went wrong. Mostly fixes otherwise. Source: https://github.com/crewAIInc/crewAI/releases/tag/1.15.28
- Listed, not ranked: [Mastra core 1.77.0](https://agentpick.space/tool/mastra/) (released 2026-10-14, Tested: no): A strong pick for TypeScript teams that need agents to run code in sandboxes. The new Factory pieces matter most if you run many sandboxed sessions. Source: https://github.com/mastra-ai/mastra/releases/tag/%40mastra/core%401.77.0
- Listed, not ranked: [Flocks v2026.10.13](https://agentpick.space/tool/flocks/) (released 2026-10-13, Tested: no): Worth a look if you're building multi-agent workflows for security operations. Its README positions it as a SecOps platform, not a general-purpose framework. We checked the repo and release notes but didn't install it (it needs uv, Node 22, and agent-browser). Source: https://github.com/AgentFlocks/flocks/releases
- Listed, not ranked: [Google Agent Development Kit (Python) 2.12.0](https://agentpick.space/tool/google-adk/) (released 2026-10-13, Tested: no): A solid release if you build on Google's stack: human-in-the-loop approvals and tool-output filtering are practical safety controls. The 40% speed-up is Google's own claim; we haven't measured it. Source: https://github.com/google/adk-python/releases/tag/v2.12.0
- Listed, not ranked: [Agno 3.1.3](https://agentpick.space/tool/agno/) (released 2026-10-12, Tested: no): Interesting if you want to wrap the Claude and Codex agent SDKs in a durable, multi-replica service. Niche for simple single-process agents. Source: https://github.com/agno-agi/agno/releases/tag/v3.1.3
- Listed, not ranked: [LangGraph 1.2.15](https://agentpick.space/tool/langgraph/) (released 2026-10-12, Tested: no): The low-level choice when you need explicit control over agent state, branching and persistence. More code than CrewAI's role-based crews, but more control. Source: https://github.com/langchain-ai/langgraph
- Listed, not ranked: [Microsoft Agent Framework (Python) 1.21.0](https://agentpick.space/tool/ms-agent-framework/) (released 2026-10-08, Tested: no): Most relevant for teams in the Microsoft/Azure stack that need policy controls such as Purview. Otherwise this is an incremental release. Source: https://github.com/microsoft/agent-framework/releases/tag/python-1.21.0
- Listed, not ranked: [OpenAI Agents SDK (Python) 0.23.1](https://agentpick.space/tool/openai-agents-sdk/) (released 2026-10-02, Tested: no): A patch release with no new features. Listed for completeness; update when convenient. Source: https://github.com/openai/openai-agents-python/releases/tag/v0.23.1

### [MCP servers](https://agentpick.space/category/mcp-servers/)
Tools and data your agents can plug into.
- Listed, not ranked: [Context7 MCP 4.4.1](https://agentpick.space/tool/context7/) (released 2026-10-13, Tested: no): Useful for cutting hallucinated APIs in coding agents, since it feeds current docs into context. Bundle packaging makes installation simpler. Source: https://github.com/upstash/context7/releases/tag/mcpb-v4.4.1
- Listed, not ranked: [MCP reference servers 1.0.0](https://agentpick.space/tool/mcp-reference-servers/) (released 2026-10-11, Tested: no): The 1.0.0 mark is a good signal for teams that held off on the reference servers. Start with filesystem and memory. Source: https://github.com/modelcontextprotocol/servers/releases/tag/v1.0.0
- Listed, not ranked: [GitHub MCP Server 2.0.2](https://agentpick.space/tool/github-mcp-server/) (released 2026-10-08, Tested: no): The default way to give agents GitHub access. This patch matters if you call it over HTTP. Source: https://github.com/github/github-mcp-server/releases/tag/v2.0.2

### [No-code automation](https://agentpick.space/category/automation/)
Workflows without code.
- Nothing new this week.

### [Research & deep search](https://agentpick.space/category/research/)
Agents that find and synthesise.
- Nothing new this week.

### [Video](https://agentpick.space/category/video/)
Generation & editing.
- Listed, not ranked: [HeyGen Video (on Civitai)](https://agentpick.space/tool/heygen-video/) (released 2026-10-13, Tested: no): A good pick for short social clips with sound built in. The duration cap makes it a clip tool, not a storytelling tool. Source: https://civitai.com/articles/36653/heygen-video-is-now-on-civitai
- Listed, not ranked: [Synthesia Syren (beta)](https://agentpick.space/tool/synthesia-syren/) (released 2026-10-09, Tested: no): The clearest "agent as the editor" take on business video so far. It's aimed at explainers and marketing, not cinematic work. Source: https://www.synthesia.io/post/syren-studio
- Listed, not ranked: [Vidu Q4 Preview](https://agentpick.space/tool/vidu-q4/) (released 2026-10-07, Tested: no): High-end quality at a low per-second price for creators who need reference consistency. It's still a preview. Source: https://www.unite.ai/vidu-releases-q4-preview-of-next-generation-flagship-ai-video-model/
- Listed, not ranked: [HiDream vivago R1](https://agentpick.space/tool/vivago-r1/) (released 2026-10-07, Tested: no): Interesting for agent pipelines thanks to the CLI. Its planned HiDream-O1-Video model integration hasn't shipped yet. Source: https://www2.newsfilecorp.com/release/317877/HiDream.ai-Makes-vivago-R1-Available-Globally-for-EndtoEnd-AI-Video-Production

### [Image & design](https://agentpick.space/category/image-design/)
Visuals and UI.
- Nothing new this week.

### [Voice & audio](https://agentpick.space/category/voice-audio/)
Speech, music, sound.
- Nothing new this week.

### [Browser / computer-use](https://agentpick.space/category/browser-agents/)
Agents that click for you.
- Listed, not ranked: [Browser Use 0.13.11](https://agentpick.space/tool/browser-use/) (released 2026-10-07, Tested: no): Worth watching for Claude users, but per its own release notes the new integration can't be used with today's public Anthropic SDK. Source: https://github.com/browser-use/browser-use/releases/tag/0.13.11
- Listed, not ranked: [Playwright MCP 0.0.83](https://agentpick.space/tool/playwright-mcp/) (released 2026-09-28, Tested: no): A reliable, scriptable way to give any MCP-capable agent a browser. A good pairing with coding agents for end-to-end checks. Source: https://github.com/microsoft/playwright-mcp/releases/tag/v0.0.83

### [Writing & marketing](https://agentpick.space/category/writing-marketing/)
Content and copy.
- Nothing new this week.

### [Support & sales agents](https://agentpick.space/category/support-sales/)
Customer-facing agents.
- Listed, not ranked: [Siena Social Agent](https://agentpick.space/tool/siena-social-agent/) (released 2026-10-13, Tested: no): It's for e-commerce brands, not solo builders. Source: https://www.siena.cx/blog/siena-social-agent-social-media-management

### [Data & analytics](https://agentpick.space/category/data-analytics/)
Agents for numbers.
- Nothing new this week.