Quickstart
Go from zero to your first API call in under 5 minutes.
1. Create an account
Sign up at mindgraph.cloud/login. The free tier includes 200 credits per month with unlimited queries and API access.
2. Get an API key
Go to Dashboard → API Keys and create a new key. Copy it — you won't be able to see it again.
3. Install an SDK
npm install mindgraph
4. Make your first call
import { MindGraph } from "mindgraph"
const graph = new MindGraph({
baseUrl: "https://api.mindgraph.cloud",
apiKey: "mg_your_key_here",
})
// Add a node
const node = await graph.addNode({
label: "User prefers dark mode",
node_type: "Preference",
})
console.log("Created:", node.uid)
// Search for it
const results = await graph.search("dark mode")
console.log("Found:", results.length, "results")
// Connect to another node
await graph.addLink({
from_uid: node.uid,
to_uid: "some_user_uid",
edge_type: "BelongsTo",
})
5. Complete example
A realistic workflow: create an entity, make a claim about it, build an argument with evidence, search for it, and traverse the reasoning chain.
import { MindGraph } from "mindgraph"
const graph = new MindGraph({
baseUrl: "https://api.mindgraph.cloud",
apiKey: "mg_your_key_here",
})
// 1. Create entities using first-class type methods (preferred)
const org = await graph.findOrCreateOrganization("Acme Corp", { description: "SaaS startup" })
console.log("Organization:", org.uid)
// findOrCreateEntity still works as a fallback for other types
const entity = await graph.findOrCreateEntity("TypeScript", { entity_type: "language" })
console.log("Entity:", entity.uid)
// 2. Build an argument with claim + evidence
const arg = await graph.argue({
claim: {
label: "TypeScript adoption",
props: { content: "TypeScript improves developer productivity" },
},
evidence: [
{ label: "Type safety", props: { description: "Static types catch bugs at compile time" } },
{ label: "IDE support", props: { description: "Rich autocomplete and refactoring tools" } },
],
warrant: {
label: "Type system principle",
props: { principle: "Static analysis reduces runtime errors" },
},
agent_id: "my-agent",
})
console.log("Claim:", arg.claim_uid)
// 3. Search for it
const results = await graph.search("TypeScript productivity")
console.log("Found:", results.length, "results")
// 4. Traverse the reasoning chain
const chain = await graph.reasoningChain(arg.claim_uid, 3)
console.log("Chain:", chain.map(s => s.label).join(" -> "))
Next steps
- Concepts — Learn about the six cognitive layers, node types, and edge types.
- Cognitive Endpoints — Full reference for all 18 cognitive endpoints with action enums and request/response schemas.
- API Reference — Full reference for all 77 endpoints.
- SDKs — Detailed guides for TypeScript and Python.