MCP tool reference · #452 of 563 by publishers
chat
Exposed by 14 servers from 8 publishers, read from the live servers by the Serverdex check; most of them sit in AI & Agents; 64.3% of them take a max_tokens parameter; the most-starred is Send Email.
chat(max_tokens?: integer, messages: array, model: string, temperature?: number, system?: string, …)Parameters
share = % of the 14 servers whose schema has it · required = % of those that mark it required| Parameter | Type | Share | Required | Usual description |
|---|---|---|---|---|
| max_tokens | integer | 64.3% | 0% | Cap on completion tokens. |
| messages | array | 57.1% | 75% | The conversation, OpenAI chat format. |
| model | string | 57.1% | 87.5% | A chat model id from list_models. |
| temperature | number | 57.1% | 0% | 0 to 2, default 0.6; lower is more predictable |
| system | string | 50% | 0% | Instructions placed ahead of the conversation as a system message. |
| image_urls | array | 35.7% | 0% | Image URLs attached to the last user message, for the chat models that take images (input_modalities in list_models). A file generate made works. |
| web_search | boolean | 35.7% | 0% | True: the model may search the web while it answers (billed per search, the price in GET /v1/catalog server_tools). |
| message | string | 28.6% | 100% | El mensaje para el agente |
| chat_id | string | 21.4% | 0% | — |
| prompt | string | 21.4% | 33.3% | A single user message (use this or messages), up to 24,000 characters |
| domain | string | 7.1% | 0% | Asset domain override — bypasses the auto-classifier. |
| include_facts | boolean | 7.1% | 0% | When true, attach fact_projection (neutral Fact[] + thesis) to the result. Default off. |
Source: the input schema each server returns for chat, aggregated over the 14 index-tier servers; type and description are the values most servers use. Rebuilt every 6 hours.
Servers with chat
12 of 14 · 8 publishers · 12 answering live · most-starred first, 3 per publisher| # | Server | Publisher | Category | ★ | Live | How it describes chat |
|---|---|---|---|---|---|---|
| 1 | Send Emailhal9ai · Communication | hal9ai | Communication | — | yes | Send emails from natural language prompts. Hosted at https://cloudish.ai. |
| 2 | Run Pythonhal9ai · Developer Tools | hal9ai | Developer Tools | — | yes | Run Python and install PyPI packages from natural language. Hosted at https://cloudish.ai. |
| 3 | Fetch URLhal9ai · Search & Web | hal9ai | Search & Web | — | yes | Fetch and extract the text content of any URL. Hosted at https://cloudish.ai. |
| 4 | PentestLLMuplinkswitch · Security & Identity | uplinkswitch | Security & Identity | — | yes | Sends messages to any chat model id through POST /v1/chat/completions and returns the answer, the token usage and what the request cost. system adds instructions, image_urls lets a model that sees images look at them (a file generate made works), web_search lets the model search while it answers. Billed to the key in the MCP request's Authorization header (Bearer sk-ptl-live-...); without one it f |
| 5 | Unrestrictableuplinkswitch · AI & Agents | uplinkswitch | AI & Agents | — | yes | Sends messages to any chat model id through POST /v1/chat/completions and returns the answer, the token usage and what the request cost. system adds instructions, image_urls lets a model that sees images look at them (a file generate made works), web_search lets the model search while it answers. Billed to the key in the MCP request's Authorization header (Bearer sk-unr-live-...); without one it f |
| 6 | RelaySwitchuplinkswitch · Developer Tools | uplinkswitch | Developer Tools | — | yes | Sends messages to any chat model id through POST /v1/chat/completions and returns the answer, the token usage and what the request cost. system adds instructions, image_urls lets a model that sees images look at them (a file generate made works), web_search lets the model search while it answers. Billed to the key in the MCP request's Authorization header (Bearer sk-relay-live-...); without one it |
| 7 | UnoRouterunorouter · AI & Agents | unorouter | AI & Agents | — | yes | Send a single prompt to any UnoRouter model and return the reply. Use search_models to find model IDs; :free models cost nothing (light per-model rate limit, rotate models if you hit 429). |
| 8 | SingIt x402singitai · Search & Web | singitai | Search & Web | — | yes | Paid per call from your key's SingIt balance and owner-set limits in USDC. The server enforces per-request and daily caps; a call over a cap is refused before anything is charged. Use list_models for model ids and prices, get_allowance (or get_limits) for what remains and which cap would refuse a call, and get_receipts for the charge. Pass max_tokens high enough for reasoning models: thinking uses |
| 9 | entropiav2luis-cwk · Crypto & Web3 | luis-cwk | Crypto & Web3 | — | yes | Conversacion con el agente creativo entropiav2. Especializado en blockchain, arte generativo, danza y contratos inteligentes. |
| 10 | SiteCheckbck-stack · Crypto & Web3 | bck-stack | Crypto & Web3 | — | yes | LLM chat completion with Llama 3.3 70B (fast): send messages or a prompt, up to 1,024 output tokens. Price: $0.004 per successful call, paid with x402 (USDC on Base, Solana and 10 more networks); failed calls are not charged. |
| 11 | JarvisClaw — AI Gateway & API Marketplaceapi-jarvisclaw · APIs & Integrations | api-jarvisclaw | APIs & Integrations | — | yes | Send a chat completion to any model on JarvisClaw through one OpenAI-compatible endpoint — GPT, Claude, Gemini, DeepSeek, and more. Pass a specific model ID or use 'auto' for smart routing to the best fit. One tool, every frontier model, no per-provider setup. |
| 12 | Marketintellravidsrk · Finance | ravidsrk | Finance | — | yes | Run a query through MarketIntell's brain (auto-routes to T1 fast agent or T2 CoT pipeline based on intent). Returns synthesized analysis with citations and trade setup when relevant. |
Stars from GitHub as stored in the Serverdex index (re-read every 7 days per server); live = answered the last daily check of its hosted endpoint. The other 2 servers are reached from their publisher pages.