Integration

TypeSafe integration

TypeSafe answers typed questions about any content with calibrated probabilities instead of generated text: send a state (text, a JSON object or an array) and named questions of four types (noul for yes/no, choice to pick one option, score to rate on an ordered rubric, box to locate something in an image) and get one typed answer per question with its probabilities and confidence, for classification, triage, routing and guardrails. List the models and aliases your account can use. Billing is per input token (output tokens are free). The multipart image upload form is not exposed (images go as base64 assets). Authentication is a TypeSafe API key, created in the TypeSafe console, sent as a Bearer token in the Authorization header.

Connect TypeSafe How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in TypeSafe.Built from the 2 TypeSafe tools below, on the connection you add once.

What you can do with TypeSafe

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Model, 1 operation

list_modelsList modelsGETAgents: ask firstLists the model names and aliases your account can send as model in ask_questions (for example jev-latest), each with a description and release date. Versioned ids such as jev-1.14.0 are accepted even when not listed.

Question, 1 operation

ask_questionsAsk questionsPOSTAgents: ask firstAsks named questions about one state and returns a typed answer per name: noul (yes/no probability), choice (one option), score (ordered rubric) or box (image location), with probabilities and confidence. Billed per input token.

Use TypeSafe from Claude Code, Cursor or Codex

Connect TypeSafe in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its TypeSafe operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect TypeSafe

Create your workspace
Sign up free. Nothing to install.
Connect TypeSafe once
Add your TypeSafe connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.