Today's drop: 1 new tool, 3 listing updates
The directory lists 150 hand-checked tools. New today: OpenBot, a free, open-source workspace from CopilotKit where AI coworkers get their own browser, files and tools, ask before they act and keep an audit trail, on your own Mac, Windows PC or servers. Updated: Claude (Sheets, Whiteboards and Data), ElevenLabs (Scribe v3) and AI-Infra-Guard (v4.6.10). Agents & Automation grows to 11 tools; we retired one Writing listing to stay at 150. Pricing only where the official page states it.
Previous drop: Nov 11, 2026 →
Get the weekly dropThe best new AI tools, one email a week. Free.
New in the directory 1
Browse all 150Tool
Category
Price
Status
Link
OpenBot (CopilotKit)Free, open-source (MIT) self-hostable workspace for AI coworkers from CopilotKit, the team behind the AG-UI agent protocol. Create a Bot with instructions and tools, pick a model (OpenAI, Anthropic, Google, xAI or any OpenAI-compatible endpoint, including a local model in Ollama or vLLM), and it works on a computer of its own with a browser, files and tools, asking for approval before actions and keeping an audit trail after. Bring agents you already built with LangGraph, Google ADK, the Claude Agent SDK and other frameworks through AG-UI; tools connect over MCP. The desktop app for Mac (Apple silicon and Intel) and Windows installs the whole workspace, backend included (it runs on Docker or Podman), with no terminal needed; teams can deploy it on Kubernetes. It depends on CopilotKit Intelligence, which is licensed separately: a free local license for personal use (desktop setup handles it), a free trial license for Kubernetes evaluation and a paid production license for teams. Optional skins copy the look of OpenAI dots, Meta Muse and Claude Duet. Web, iOS, Android, Slack and Teams access are early access only. OpenBot 1.0 launched Nov 10, 2026 (v1.0.2 on Nov 11); about 11,200 GitHub stars. Very new, so give each Bot only the tools and permissions it needs.
Price Free for personal use (MIT app plus a free local Intelligence license); you pay your model provider. OpenBot for Business needs a paid production license (see open.bot)
Why this one
- OpenBot is what you get if you want a Claude Duet or OpenAI dots style AI coworker but on your own machine and your own model keys. CopilotKit, the team behind the AG-UI agent protocol, launched 1.0 on Nov 10. You create a Bot with instructions and tools, pick a model (OpenAI, Anthropic, Google, xAI or any OpenAI-compatible endpoint, including a local model in Ollama or vLLM), and the Bot works on a computer of its own with a browser and files, asks before it acts and logs what it did. Agents built with LangGraph, Google ADK or the Claude Agent SDK can plug in over AG-UI, and tools connect over MCP. The Mac and Windows desktop app installs everything, backend included, without a terminal (it runs on Docker or Podman). The app is MIT-licensed, but it depends on CopilotKit Intelligence, which is licensed separately: free locally for personal use, paid for production team deployments. About 11,200 GitHub stars and three releases in three days (v1.0.2 on Nov 11). For IT learners, it is a hands-on way to see what approval gates, per-tool permissions and audit logs for AI agents look like, the controls a help desk or security team will ask about. Start with a narrow Bot and only the tools it needs.
Listing updates
Checked on Nov 12, 2026 against official pages and release notes.
- Claude: Claude Sheets (spreadsheets Claude builds and edits with you, with tabs, formulas, formatting and drop-downs, export to .xlsx or CSV) and Claude Whiteboards (shared boards Claude turns into a timeline with owners and dates) are in beta on all paid plans on web, desktop and mobile as of Nov 10. Claude Dashboards is now generally available as Claude Data, and Claude Meetings, which joins Google Meet calls to take notes, is in beta for Team and Enterprise. Enterprise admins have to turn Sheets and Whiteboards on.
- ElevenLabs: Scribe v3 and Scribe v3 Live (Nov 10) are its new speech-to-text models and the defaults in the app. ElevenLabs says v3 makes 27% fewer word errors than v2 across 102 languages and v3 Live 43% fewer than v2 Realtime (its own tests); v3 Live adds realtime speaker labels. API users switch with the model ID
scribe_v3orscribe_v3_live. - AI-Infra-Guard: v4.6.10 (Nov 12) adds a local generative AI safety self-check, one page that runs refusal checks and infrastructure scans and keeps a printable report without storing your API keys, plus about 135 new AI product CVE advisories from NVD and MCP-scan and agent-scan fixes. Still no login, so keep it off public networks.
Category changes
- Agents & Automation grows to 11 tools with OpenBot.
- Writing drops to 4: we retired Writesonic to stay at the 150-tool cap. It now focuses on SEO and AI-search visibility for marketing teams, has no free plan, and Jasper and Copy.ai cover the same ground for our readers. Its old page redirects to the Writing category.
- AI for IT & Cybersecurity stays at 60 tools.
Checked, not listed
- A2UI 1.0 (Google, Nov 11): the first stable version of Google's open protocol that lets agents answer with forms, charts and other native UI instead of text, with renderers for web, Android and iOS. It is a protocol for developers rather than a tool you use directly, so it waits.
- Gemini Decisions API (Google, Nov 11): a new Gemini API endpoint and model (gemini-decision-1) that scores up to 1,000 yes/no, choice or rubric questions in one pass at $0.10 per million input tokens. Useful for routing and policy checks; we are watching it.
- Databricks Genie (always-on agent): enterprise-only, and the always-on version is still coming to preview.
- Codex 0.167.2, Orca v1.4.257 and CISO Assistant v4.2.3 (evidence review mode, Canada's ITSP.10.033 framework mapped to NIST SP 800-53): routine releases, no listing change. Claude Code is still on v2.1.322, Ollama v0.40.17 and Wazuh 4.14.11 are still release candidates.