Skip to main content
OrgX connects to ChatGPT through the OpenAI Apps SDK using the OrgX MCP server. Use it when ChatGPT needs durable team memory, decision approvals, artifact recall, agent delegation, or project execution status.
As of April 27, 2026, ChatGPT support depends on whether your ChatGPT plan and workspace expose remote MCP connectors / Developer Mode. If that option is unavailable in your account, use Cursor, Claude, VS Code, or the OpenClaw plugin instead.

Questions This Page Answers

  • How do I use ChatGPT as a control surface for OrgX, not just a chatbot?
  • Which tools should I call for approvals, blockers, and new work dispatch?
  • What exact agent IDs should I use when spawning specialist tasks?

Features

Approve Decisions

Review and approve pending decisions with rich widget UI.

Agent Status

See what your agents are working on in real-time.

Org Memory

Search your knowledge graph for context and history.

Spawn Tasks

Assign new work to specialist agents.

Getting Started

1

Enable Developer Mode

  1. Open ChatGPT Settings -> Apps & Connectors
  2. Turn Developer Mode on
2

Add the OrgX Connector

  1. In Apps & Connectors, click Add Connector (or Add MCP Server)
  2. Enter the MCP endpoint: https://mcp.useorgx.com/mcp
  3. Save — OrgX will appear in your connectors list
3

Authenticate

On first use, ChatGPT will prompt you to connect your OrgX account:
  1. Click the authentication link
  2. Sign in to OrgX (creates account if needed)
  3. Grant ChatGPT permission to access your workspace
OrgX uses OAuth 2.1 for secure authentication. Your credentials are never stored by ChatGPT.
4

Start Using

Once connected, just talk to ChatGPT naturally:
  • “Show me pending decisions”
  • “What are my agents working on?”
  • “Approve the marketing campaign decision”
  • “Search for our pricing strategy decisions”
  • “Remember this decision for the team: we are moving onboarding analytics to PostHog.”

Available Commands

Tier-0 Intent Tools

Start with these wrapper tools. They map to the natural phrases ChatGPT users actually type while preserving the underlying OrgX behavior.
User intentPreferred toolUnderlying OrgX action
Remember a team decisionorgx_decideRecords a decision in organizational memory
Recall prior contextorgx_searchSearches decisions, artifacts, initiatives, and project context
Review blocked agent workorgx_decideLists pending approvals, then approves or rejects after confirmation
Hand work to an agentorgx_spawnAssigns work to a specialist OrgX agent
Check project progressorgx_recommendReturns initiative health, blockers, milestones, owners, and activity

Decision Management

What it does: Retrieves all decisions waiting for your approval.Example prompts:
  • “Show my pending decisions”
  • “What needs my approval?”
  • “Any critical decisions waiting?”
Parameters:
  • limit (optional): Maximum number to return
  • urgency_filter (optional): all, critical, or high
  • initiative_id (optional): Filter by specific initiative
Response: Rich widget showing decision cards with approve/reject buttons.
What it does: Approves a pending decision and triggers agent execution.Example prompts:
  • “Approve the homepage redesign decision”
  • “Approve decision D-123 with note: looks good”
Parameters:
  • decision_id (required): ID of the decision
  • note (optional): Comment to attach
Response: Confirmation widget showing approval status.
What it does: Rejects a decision with feedback for the agent.Example prompts:
  • “Reject the email campaign, needs more personalization”
  • “Reject D-456 because budget is too high”
Parameters:
  • decision_id (required): ID of the decision
  • reason (required): Why you’re rejecting
Response: Confirmation widget. Agent receives feedback and may revise.
What it does: Retrieves past decisions on a topic.Example prompts:
  • “What decisions have we made about pricing?”
  • “Show me past marketing decisions”
Parameters:
  • topic (required): What to search for
  • initiative_id (optional): Filter by initiative
  • limit (optional): Maximum results

Agent Management

What it does: Shows current agent activity and status.Example prompts:
  • “What are my agents doing?”
  • “Show me active agents”
  • “Is the engineering agent working on anything?”
Parameters:
  • agent_id (optional): Filter to specific agent
  • include_idle (optional): Show idle agents too
Response: Widget showing agent cards with current tasks.
What it does: Assigns a new task to a specialist agent.Example prompts:
  • “Have the product agent analyze our competitor positioning”
  • “Ask the marketing agent to draft a blog post about our new feature”
Parameters:
  • agent (required): Canonical agent ID (engineering-agent, product-agent, marketing-agent, sales-agent, design-agent, operations-agent, orchestrator-agent)
  • task (required): What to do
  • context (optional): Additional background
  • initiative_id (optional): Link to an initiative
Response: Confirmation that task was created and assigned.
This requires write permissions. Make sure you’ve granted agent write scopes during authentication.
What it does: Searches your organizational knowledge graph.Example prompts:
  • “What do we know about enterprise pricing?”
  • “Find artifacts related to the Q1 launch”
  • “Search for previous security audits”
Parameters:
  • query (required): What to search for
  • scope (optional): all, artifacts, decisions, or initiatives
  • limit (optional): Maximum results
Response: Widget showing relevant items with excerpts and links.
What it does: Shows health and progress of an initiative.Example prompts:
  • “How is the product launch initiative doing?”
  • “Show me the pulse for initiative I-789”
Parameters:
  • initiative_id (required): Which initiative to check
Response: Widget with health score, milestones, blockers, and recent activity.

Widgets

OrgX uses rich widgets to display information in ChatGPT, not just plain text:

Decisions Widget

Interactive cards with approve/reject buttons, context, and evidence links.

Agent Status Widget

Visual cards showing each agent’s current task, progress, and health.

Search Results Widget

Organized results with type icons, excerpts, and deep links.

Initiative Pulse Widget

Dashboard view with health score, milestone progress, and activity feed.

Permissions & Security

Scopes

When you authenticate, you can grant different permission levels:
ScopeWhat It Allows
decisions:readView pending and past decisions
decisions:writeApprove or reject decisions
agents:readView agent status and activity
agents:writeSpawn new agent tasks
initiatives:readView initiative details
memory:readSearch organizational knowledge
The authenticated ChatGPT connector uses https://mcp.useorgx.com/mcp and requires OrgX OAuth. For no-auth exploration, use the discovery-only public endpoint at https://mcp.useorgx.com/public; it returns examples and connection help, not workspace data.

Data Privacy

  • Your messages in ChatGPT are processed by OpenAI per their policies
  • OrgX only receives the specific tool calls, not your full conversation
  • All API calls are authenticated and scoped to your workspace
  • We never store your ChatGPT conversation history

Troubleshooting

During the current preview, connectors are private and only appear after you add them in Developer Mode.
  1. Verify Settings -> Apps & Connectors -> Developer Mode is enabled
  2. Confirm you’ve added https://mcp.useorgx.com/mcp as a connector
  3. Restart ChatGPT if the list doesn’t refresh
  1. Try clearing your browser cache and re-authenticating 2. Make sure you’re signing into the correct OrgX workspace 3. Check that your OrgX account is in good standing
  • Ensure you have pending decisions in your OrgX workspace - Check that you’ve granted decisions:read permission - Try refreshing by asking “Show pending decisions” again
Write operations require the decisions:write scope. Re-authenticate and make sure to grant write permissions.

Next Steps

MCP Client Setup

Connect OrgX from Cursor, Claude, VS Code, and other MCP clients.

MCP Tools Reference

Full API documentation for all tools.