Custom AI Setup

Connect any AI assistant via OpenAPI, REST, or MCP.

Connection endpoints

AiM Continuity API base URL

/api

OpenAPI schema URL

/api/openapi.json

MCP server URL

/mcp

Copy these into your Custom AI configuration. The actual API key is never shown here.

Supported connection methods

REST / OpenAPI ActionRemote MCP

Setup steps

1.Fetch the AiM OpenAPI schema to generate client bindings.
2.Choose REST or remote MCP as the transport.
3.Send the Bearer credential in the Authorization header on every request.
4.Confirm the assistant can call read and recommend tools only.
5.Run the connection test.

Authentication

·Authentication: Bearer token via the Authorization header.
·REST endpoints and the remote MCP server share one intelligence layer.
·Use the developer documentation to map your assistant's tool format.

Example request

http

POST /api/continuity/intake
Authorization: Bearer YOUR_AIM_CREDENTIAL
Content-Type: application/json

{
  "provider": "custom",
  "conversation": { "title": "Strategy sync" },
  "organization": { "primary_project": "Website Redesign" },
  "tasks": [{ "title": "Follow up with design", "status": "OPEN" }],
  "recommendations": [{ "action": "ASSIGN_TO_PROJECT", "confidence": 0.8 }]
}

Custom AI setup prompt

Paste this into Custom AI's system / instructions so it knows how to push conversations, decisions, tasks, and ideas into AiM Continuity as PENDING REVIEW.

prompt

You are connected to AiM Continuity, my work-intelligence layer.

When I describe or reference work — a conversation, decision, task, idea, or
asset — submit it to AiM using POST /api/continuity/intake with:

{
  "provider": "custom",
  "source": "custom_api",
  "external_conversation_id": "<current conversation id or slug>",
  "conversation": { "title": "<title>", "summary": "<summary>", "date": "<ISO date>" },
  "organization": {
    "primary_project": "<project>",
    "workstream": "<workstream>",
    "related_projects": ["<related>"],
    "brands": ["<brands>"],
    "websites": ["<urls>"]
  },
  "decisions": [{ "title": "<decision>", "status": "PROPOSED" }],
  "tasks": [{ "title": "<task>", "priority": "HIGH|MED|LOW", "status": "OPEN" }],
  "ideas": [{ "title": "<idea>" }],
  "assets": [{ "name": "<name>", "type": "url", "url": "<url>" }],
  "recommendations": [{ "action": "ASSIGN_TO_PROJECT", "reason": "<why>", "confidence": 0.8 }]
}

Before submitting, call GET /api/search?q=<topic> and GET /api/projects to
check for existing records. If a record exists, recommend UPDATE EXISTING.
All submissions land as PENDING REVIEW for a human to approve. AI Recommends.
Humans Decide.

First connection test

After connecting, have Custom AI run these two prompts. The second must land as PENDING REVIEW — never auto-approved.

1. “Show me my active AiM Continuity projects.”

2. “Recommend adding a task called Test AiM Connection to my selected project.”

Guided setup

Prefer a guided flow? Use the connection wizard to choose a method, create a credential, test, choose permissions, and finish — in six steps.

Open the wizard

Principle

AI Recommends. Humans Decide. Custom AI may read organizational records and submit recommendations, but may never delete, approve its own work, or change configuration.