Automation & agents
48 scored tools, ranked by overall score. Last checked on 25 October 2026. Refine with the Finder →
The three to remember
The best free agent with your own key; it executes commands, keep an eye on API costs.
- · Utility 9/10, best of the selection
- · Free, no limits
- · One security caveat, see the card
Official and cautious by design; limited to paid plans.
- · Utility 7/10
- · Paid, from €18
- · No known security caveat
The most serious alternative to OpenClaw, cleaner on privacy (MIT, zero telemetry), and able to import your OpenClaw config.
- · Utility 8.5/10
- · Free tier to get started
- · No known security caveat
| Tool | Price | Score | Sec. |
|---|---|---|---|
| Hermes Agent (Nous Research)RisingSelf-hosted agent that learns from its tasks, creates its own skills, maintains persistent memory, and responds on Telegram, Discord, Slack, WhatsApp, or Signal. | Freemium | 9.1 | Security: nothing to flag |
| CrewAIRisingPython framework for orchestrating teams of agents by roles. | Freemium | 8.8 | Security: nothing to flag |
| LangGraphFramework for building stateful agents with an explicit control graph. | Freemium | 8.8 | Security: nothing to flag |
| Actions ComposioRisingMore than 1,000 action toolkits for SaaS services, with managed OAuth. | Freemium | 8.4 | Security: one caveat, see the card |
| Arcade.devRising7,500 tools across 81 MCP servers, with per-call OAuth and tokens never exposed to the model. | Freemium | 8.4 | Security: one caveat, see the card |
| Claude Agent SDKNewThe engine behind Claude Code made available to developers: files, terminal, web search, subagents, skills, and fine-grained permissions in your own agents. | Freemium | 8.1 | Security: one caveat, see the card |
| Composio (connect-apps)RisingGives the agent actions across 1,000+ applications (Gmail, Slack, Notion, GitHub…) via centralized authentication. | Freemium | 7.5 | Security: one caveat, see the card |
| Zapier MCP et AI Actions30,000 actions across 9,000 apps behind a single MCP URL. | Freemium · from 18 € | 7.5 | Security: one caveat, see the card |
| Harpa AIWeb assistant and automation in the browser, with your own API key. | Freemium · from 12 € | 7.1 | Security: one caveat, see the card |
| Nœuds outils n8nAny n8n node exposed as a tool to an agent. | Freemium | 7.1 | Security: one caveat, see the card |
| Pipedream Connect10,000 tools across 2,000+ applications with managed OAuth. | Freemium | 7.1 | Security: one caveat, see the card |
| Meta MuseNewMeta's personal agent: it connects to your emails, calendar, payments and apps to act on your behalf, from an app, WhatsApp or a Mac. | Freemium | 7.0 | Security: one caveat, see the card |
| Browserbase et StagehandRisingHosted Chrome browsers and act, extract, observe SDK for agents. | Freemium · from 18 € | 6.7 | Security: serious reservations |
| Browser UseRisingLLM-driven browser control, open source. | Freemium | 6.7 | Security: serious reservations |
| Outils d'agenda (Google, Outlook)Create and reschedule events. | Freemium | 6.7 | Security: one caveat, see the card |
| Outils SlackRead channels and post messages. | Freemium | 6.7 | Security: one caveat, see the card |
| Actors Apify comme outils7,000+ scrapers and automations callable by an agent. | Freemium | 6.5 | Security: one caveat, see the card |
| Outils Gmail (Composio, Arcade, Zapier)Read, sort and send emails from an agent. | Freemium | 6.5 | Security: serious reservations |
| BardeenBrowser automations powered by AI: extract, fill, and send data to your tools. | Freemium | 6.2 | Security: one caveat, see the card |
App : A product you use directly · Tool : A precise action the AI can call · MCP : Standardises the connection between an AI application and systems that expose tools or data · Built-in mode : A feature to switch on inside an assistant you already use (ChatGPT, Claude, Gemini): nothing to install · Skill : A reusable method: instructions, procedures, sometimes scripts or templates · Plugin : A set of capabilities installed into an AI application or a host
Method
Overall score = utility (rated by the editorial team), reliability (GitHub activity, official publisher, reachable site), security (where data goes, permissions, known incidents) and accessibility (install, required level, price), weighted per category. A tool inactive for over six months is flagged and leaves the recommendations. No ranking is sold.
Missing a tool? A score looks wrong?
Write to the editors