LIVE
Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|
AI tools · Tool

Tools: the actions your AI can call

A tool performs a precise action for the AI: read a file, query a database, run a computation, search the web. These are the bricks agents assemble.

80 scored tool. Last checked on 25 October 2026.

The three to remember

Most powerful
Actions Composio
ToolRisingSecurity: one caveat, see the card
8.4

The largest library; centralized OAuth concentrates risk, a critical CVE in 2025 serves as a reminder.

  • · Utility 9/10, best of the selection
  • · Free tier to get started
  • · One security caveat, see the card
Freemium5 min
Simplest
Read, Grep, Glob (Claude Code)
ToolSecurity: nothing to flag
8.3

The three safe tools in Claude Code, to be permanently allowed.

  • · Utility 8/10
  • · Free, no limits
  • · No known security caveat
Free1 click
Best free
Docling (IBM)
ToolRisingSecurity: nothing to flag
9.0

The best free option: everything stays on your machine.

  • · Utility 8/10
  • · Free, no limits
  • · No known security caveat
Free5 min

For anthropic-api (8)

ToolPriceScoreSec.
tool_search (Anthropic)Lets Claude find the right tool among hundreds without loading all schemas upfront.Free7.7Security: nothing to flag
web_search (Anthropic)Server-side web search by Anthropic, with citations, enabled by a parameter in the API call.Paid · from 9 €7.3Security: one caveat, see the card
code_execution (Anthropic)Runs Python and bash in an Anthropic sandbox, with a file editor.Freemium6.9Security: one caveat, see the card
web_fetch (Anthropic)Fetches and reads a page or PDF at a given URL.Paid6.9Security: one caveat, see the card
text_editor (Anthropic)View, create, and replace text in files, client-side.Free6.6Security: one caveat, see the card
memory (Anthropic)Persistent Claude memory stored in files you host.Free6.5Security: one caveat, see the card
bash (Anthropic, côté client)Claude requests shell commands that your code executes.Free6.4Security: serious reservations
computer use (Anthropic)Lets Claude see the screen, move the mouse, and type on the keyboard on a machine.Paid5.3Security: serious reservations

For Claude Code (6)

ToolPriceScoreSec.
Read, Grep, Glob (Claude Code)File reading and project search, without writing.Free8.3Security: nothing to flag
Permissions (Claude Code)Allow, deny, and ask rules per tool and per pattern.Free8.2Security: nothing to flag
Edit, Write (Claude Code)Modifying and creating files in the project.Free7.7Security: one caveat, see the card
Bash (Claude Code)Executes shell commands in a persistent session.Free7.1Security: serious reservations
Task, sous-agents (Claude Code)Launches an autonomous sub-agent with its own tools.Free6.9Security: one caveat, see the card
WebFetch, WebSearch (Claude Code)Web search and page retrieval from the code agent.Free6.6Security: one caveat, see the card

For openai-api (6)

ToolPriceScoreSec.
file_search (OpenAI)Search your documents indexed in an OpenAI vector store.Paid · from 2 €8.0Security: nothing to flag
image_generation (OpenAI)Image generation and editing callable by an agent.Paid7.4Security: nothing to flag
web_search (OpenAI)Web search hosted within the Responses API.Paid · from 9 €7.3Security: one caveat, see the card
code_interpreter (OpenAI)Executes Python in an OpenAI container, with file input and output.Paid6.8Security: one caveat, see the card
shell (OpenAI)Shell hosted in a Debian container, network off by default.Paid5.6Security: serious reservations
computer_use (OpenAI)Control of a browser or system by the model.Paid5.3Security: serious reservations

For composio (4)

ToolPriceScoreSec.
Actions ComposioRisingMore than 1,000 action toolkits for SaaS services, with managed OAuth.Freemium8.4Security: one caveat, see the card
Outils d'agenda (Google, Outlook)Create and reschedule events.Freemium6.7Security: one caveat, see the card
Outils SlackRead channels and post messages.Freemium6.7Security: one caveat, see the card
Outils Gmail (Composio, Arcade, Zapier)Read, sort and send emails from an agent.Freemium6.5Security: serious reservations

For mistral-api (3)

ToolPriceScoreSec.
OCR (Mistral)Structured reading of scanned documents in 170 languages, with positions and confidence scores.Paid · from 4 €8.0Security: nothing to flag
code_interpreter (Mistral)Python sandbox for Mistral agents.Paid · from 28 €6.7Security: one caveat, see the card
web_search (Mistral)Web search for Mistral agents, premium version with AFP and AP.Paid · from 28 €6.5Security: one caveat, see the card

For gemini-api (3)

ToolPriceScoreSec.
google_search (Gemini)Grounds Gemini responses on Google Search, with sources.Freemium7.4Security: one caveat, see the card
code_execution (Gemini)Python executed on Google's side, results returned to the model.Freemium7.0Security: one caveat, see the card
url_context (Gemini)Reads the content of URLs you pass in the prompt.Freemium6.6Security: one caveat, see the card

For opensource (2)

ToolPriceScoreSec.
Docling (IBM)RisingLocal PDF and document parsing, tables and layout included.Free9.0Security: nothing to flag
Crawl4AIRisingOpen source crawler optimized for LLMs, running locally.Free7.1Security: one caveat, see the card

For Any agent (2)

ToolPriceScoreSec.
agent-browser (Vercel)Command-line driven browser for AI agents: accessibility snapshots with stable references (@e1), click, type, and screenshot.Free8.1Security: one caveat, see the card
Agent ReachGives an agent read and search access to the web, YouTube, Reddit, X, GitHub, LinkedIn, and a dozen other platforms, without a paid API key.Free7.6Security: one caveat, see the card

For llamaindex (2)

ToolPriceScoreSec.
LlamaParseComplex document parsing for RAG.Freemium7.7Security: nothing to flag
Tools LlamaIndexLlamaHub tools and loaders for agents and RAG.Free6.5Security: one caveat, see the card

For divers (2)

ToolPriceScoreSec.
Génération d'images, audio, vidéo comme outilsGeneration APIs called by an agent within a workflow.Paid6.8Security: nothing to flag
Outils cloud et Kubernetes (AWS, GCP, Azure, Terraform)Control infrastructure from an agent.Free5.7Security: serious reservations

For arcade (1)

ToolPriceScoreSec.
Arcade.devRising7,500 tools across 81 MCP servers, with per-call OAuth and tokens never exposed to the model.Freemium8.4Security: one caveat, see the card

For frameworks (1)

ToolPriceScoreSec.
Approbation humaine (human-in-the-loop)Pauses the agent and requests confirmation before a sensitive action.Free8.2Security: nothing to flag

For firecrawl-api (1)

ToolPriceScoreSec.
Firecrawl scrape, crawl, searchRisingFull website extraction to markdown, callable by an agent.Freemium8.1Security: one caveat, see the card

For lakera-api (1)

ToolPriceScoreSec.
Lakera GuardDetects direct and indirect prompt injections before the model or tool is called.Paid8.0Security: nothing to flag

For codex (1)

ToolPriceScoreSec.
shell, apply_patch (Codex)RisingAtomic commands and patches, under dual sandbox and approval control.Free7.9Security: serious reservations

For wolfram (1)

ToolPriceScoreSec.
Wolfram AlphaSymbolic computation and factual data callable by an agent.Freemium7.7Security: nothing to flag

For langchain (1)

ToolPriceScoreSec.
Tools et toolkits LangChainRisingOver 300 integrations usable as tools by a LangChain agent.Free7.7Security: one caveat, see the card

For vercel (1)

ToolPriceScoreSec.
Tools Vercel AI SDKTyped tool definitions for agents in TypeScript.Free7.7Security: nothing to flag

For unstructured (1)

ToolPriceScoreSec.
UnstructuredDocument chunking and extraction across all formats.Freemium7.7Security: nothing to flag

For brave-api (1)

ToolPriceScoreSec.
Brave Search APIIndependent index, no tracking.Paid · from 5 €7.6Security: nothing to flag

For prompt-security (1)

ToolPriceScoreSec.
Prompt SecurityManaged guardrail for applications and agents.Paid7.6Security: nothing to flag

For zapier (1)

ToolPriceScoreSec.
Zapier MCP et AI Actions30,000 actions across 9,000 apps behind a single MCP URL.Freemium · from 18 €7.5Security: one caveat, see the card

For protectai (1)

ToolPriceScoreSec.
LLM Guard (Protect AI)Watch outOpen source input and output scanners: injection, personal data, toxicity.Free7.4Security: nothing to flag

For mcp:github (1)

ToolPriceScoreSec.
Outils du GitHub MCPIssues, pull requests, code and security alerts, in read and write mode.Free7.3Security: one caveat, see the card

For Cursor (1)

ToolPriceScoreSec.
Outils de l'agent CursorRead, edit, terminal and search from the Cursor agent.Freemium · from 20 €7.3Security: one caveat, see the card

For e2b (1)

ToolPriceScoreSec.
E2B Code InterpreterRisingIsolated micro-VM for executing code produced by an agent.Freemium7.3Security: one caveat, see the card

For gemini-cli (1)

ToolPriceScoreSec.
run_shell_command, write_file, google_web_search (Gemini CLI)RisingThe built-in tools of Google's CLI, each with confirmation.Free7.2Security: one caveat, see the card

For pipedream (1)

ToolPriceScoreSec.
Pipedream Connect10,000 tools across 2,000+ applications with managed OAuth.Freemium7.1Security: one caveat, see the card

For n8n (1)

ToolPriceScoreSec.
Nœuds outils n8nAny n8n node exposed as a tool to an agent.Freemium7.1Security: one caveat, see the card

For openai (1)

ToolPriceScoreSec.
Tools OpenAI Agents SDKWebSearchTool, FileSearchTool, ComputerTool and your own functions.Free6.9Security: one caveat, see the card

For stripe (1)

ToolPriceScoreSec.
Stripe Agent ToolkitRisingAllows an agent to create payments, invoices, and subscriptions in Stripe.Free6.9Security: serious reservations

For anthropic (1)

ToolPriceScoreSec.
Tools Claude Agent SDKClaude Code tools exposed in your own agent, with the permissions system.Free6.9Security: one caveat, see the card

For exa-api (1)

ToolPriceScoreSec.
Exa SearchEmbedding-based semantic search, designed for agents.Freemium6.9Security: one caveat, see the card

For tavily-api (1)

ToolPriceScoreSec.
Tavily SearchWeb search and extraction designed for agents.Freemium6.9Security: one caveat, see the card

For browserbase (1)

ToolPriceScoreSec.
Browserbase et StagehandRisingHosted Chrome browsers and act, extract, observe SDK for agents.Freemium · from 18 €6.7Security: serious reservations

For browser-use (1)

ToolPriceScoreSec.
Browser UseRisingLLM-driven browser control, open source.Freemium6.7Security: serious reservations

For jina-api (1)

ToolPriceScoreSec.
Jina Reader et SearchConverts a URL into clean Markdown for a language model.Freemium6.7Security: one caveat, see the card

For huggingface (1)

ToolPriceScoreSec.
Tools smolagents (Hugging Face)Tools declared with a simple decorator, code execution by the agent.Free6.5Security: one caveat, see the card

For apify (1)

ToolPriceScoreSec.
Actors Apify comme outils7,000+ scrapers and automations callable by an agent.Freemium6.5Security: one caveat, see the card

For serper-api (1)

ToolPriceScoreSec.
SerperCheap Google results for agents.Freemium6.5Security: one caveat, see the card

For xai-api (1)

ToolPriceScoreSec.
x_search (xAI)Real-time search of posts and profiles on X.Paid · from 5 €6.5Security: one caveat, see the card

For crewai (1)

ToolPriceScoreSec.
Tools CrewAIReady-to-use tool catalogue for CrewAI agent teams.Free6.5Security: one caveat, see the card

For perplexity-api (1)

ToolPriceScoreSec.
Perplexity Sonar APICited, ready-to-use answers for an agent.Paid · from 5 €6.4Security: one caveat, see the card

For shopify (1)

ToolPriceScoreSec.
Outils Shopify pour agentsCatalog, orders, and payment via the agentic commerce protocol.Free6.3Security: one caveat, see the card

For daytona (1)

ToolPriceScoreSec.
DaytonaFast-starting sandboxes for agents.Freemium6.1Security: one caveat, see the card

For modal (1)

ToolPriceScoreSec.
Modal SandboxesServerless sandboxes with on-demand GPU.Freemium6.1Security: one caveat, see the card

For 1password (1)

ToolPriceScoreSec.
Secrets pour agents (1Password, Vault)Provide secrets to an agent without putting them in the prompt.Paid6.1Security: one caveat, see the card

For elevenlabs (1)

ToolPriceScoreSec.
Outils ElevenLabs Conversational AIFunction calling from an ElevenLabs voice agent.Paid5.9Security: one caveat, see the card

For twilio (1)

ToolPriceScoreSec.
Twilio ConversationRelayReal-time phone voice connected to an agent via WebSocket.Paid5.9Security: one caveat, see the card

For vapi (1)

ToolPriceScoreSec.
VapiFull-featured voice agent with telephony and function calling.Paid5.9Security: one caveat, see the card

For paypal (1)

ToolPriceScoreSec.
PayPal Agent ToolkitPayPal orders, invoices, disputes, and subscriptions for agents.Free5.7Security: serious reservations

For microsoft (1)

ToolPriceScoreSec.
Playwright comme outilDeterministic browser automation, exposed to the agent.Free5.7Security: serious reservations

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

What to know

A tool is not used on its own: it is called from an agent or assistant. Each enabled tool widens what the AI can do, and what it can break: enable only what the task needs.

Find the right one in 4 questions →