Run your AIs
189 scored tools, ranked by overall score. Last checked on 29 October 2026. Refine with the Finder →
The three to remember
The de facto standard for running models locally and connecting them to other tools.
- · Utility 9.5/10, best of the selection
- · Free, no limits
- · No known security caveat
The most installed skill on the official marketplace: one click, immediate effect on UI quality.
- · Utility 9/10
- · Free, no limits
- · No known security caveat
| Tool | Price | Score | Sec. |
|---|---|---|---|
| file_search (OpenAI)Search your documents indexed in an OpenAI vector store. | Paid · from 2 € | 8.0 | Security: nothing to flag |
| Lakera GuardDetects direct and indirect prompt injections before the model or tool is called. | Paid | 8.0 | Security: nothing to flag |
| OCR (Mistral)Structured reading of scanned documents in 170 languages, with positions and confidence scores. | Paid · from 4 € | 8.0 | Security: nothing to flag |
| RequestyHosted router with cost optimization and guardrails, team-oriented. | Paid | 7.7 | Security: nothing to flag |
| Brave Search APIIndependent index, no tracking. | Paid · from 5 € | 7.6 | Security: nothing to flag |
| Prompt SecurityManaged guardrail for applications and agents. | Paid | 7.6 | Security: 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 € | 7.3 | Security: one caveat, see the card |
| web_search (OpenAI)Web search hosted within the Responses API. | Paid · from 9 € | 7.3 | Security: one caveat, see the card |
| web_fetch (Anthropic)Fetches and reads a page or PDF at a given URL. | Paid | 6.9 | Security: one caveat, see the card |
| code_interpreter (OpenAI)Executes Python in an OpenAI container, with file input and output. | Paid | 6.8 | Security: one caveat, see the card |
| Amazon Bedrock AgentCoreNewAWS managed runtime to put an agent into production: per-session isolated execution, memory, tool gateway, identity, browser, code sandbox, and traces. | Paid | 6.5 | Security: one caveat, see the card |
| web_search (Mistral)Web search for Mistral agents, premium version with AFP and AP. | Paid · from 28 € | 6.5 | Security: one caveat, see the card |
| x_search (xAI)Real-time search of posts and profiles on X. | Paid · from 5 € | 6.5 | Security: one caveat, see the card |
| Perplexity Sonar APICited, ready-to-use answers for an agent. | Paid · from 5 € | 6.4 | Security: one caveat, see the card |
| Secrets pour agents (1Password, Vault)Provide secrets to an agent without putting them in the prompt. | Paid | 6.1 | Security: one caveat, see the card |
| Vertex AI Agent EngineNewGoogle Cloud managed runtime for deploying agents: execution, sessions, long-term memory (Memory Bank), code execution, and agent identity. | Paid | 6.1 | Security: one caveat, see the card |
| computer use (Anthropic)Lets Claude see the screen, move the mouse, and type on the keyboard on a machine. | Paid | 5.3 | Security: serious reservations |
| computer_use (OpenAI)Control of a browser or system by the model. | Paid | 5.3 | Security: 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
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