Skip to main content

Layer MCP Guide: Connect Layer to Claude in 3 Steps

Step-by-step guide to connect Layer MCP to Claude to create on-brand content at scale.

Updated today

Quickstart (2-minute check)

  1. An admin adds the Layer connector once using this URL:
    https://mcp.app.layer.ai/mcp

  2. Each team member connects and authorizes their Layer account in Claude Connectors.

  3. Start a new Claude conversation and ask: “What can I do with Layer?”

  4. If tools appear, run a workflow command such as list_workflows or execute_workflow.

For teams, an admin completes step 1 once. For personal accounts, the single user completes both step 1 and step 2.

What is Layer MCP?

Layer MCP is a Model Context Protocol (MCP) server endpoint that connects Claude to a Layer workspace, enabling Claude to orchestrate models, workflows, and creative tooling.

With Layer MCP, marketing teams can generate, iterate, and scale on-brand creative without leaving the Claude UI. For example: generate 20 Meta ad variations in a specific art style and return channel-ready files directly into the chat.

Plain-language definition: MCP is a standard that lets external services expose tools to Claude, so you can trigger real production tasks directly from a conversation.

Tools available via MCP:

  • get_instructions — Returns platform guidance, Creative Unit pricing info, and error codes

  • get_workflow_instructions — Returns the full workflow lifecycle: discover, estimate, execute, poll, display

  • get_forge_instructions — Returns the forge (direct model execution) lifecycle

  • list_workspaces — List all workspaces you belong to, including Creative Unit balances

  • get_workspace — Get details for a specific workspace

  • get_model_recommendations — Get curated model recommendations by category

  • list_models — Search and filter models by modality, capabilities, or text query

  • get_model — Get full details for a model including capabilities and prompt format

  • list_workflows — List available workflows with their inputs and outputs

  • estimate_workflow_price — Check the Creative Unit cost before running

  • execute_workflow — Run a workflow (async — returns a run_id)

  • get_workflow_run — Poll for workflow run status and results

  • cancel_workflow_run — Cancel a running workflow and release reserved CUs

  • estimate_forge_price — Check the Creative Unit cost for a generation

  • execute_forge — Run inference on a model (async — returns an inference_id)

  • get_forge_run — Poll for forge run status and generated assets

  • upload_file — Upload an image or video from a URL for use as input in workflows or forge

Setup Instructions

a. Via Claude.ai or Claude Desktop

Which steps do I need? Solo user or personal account → do both steps below. On a team → your admin does step 1 once, then each member does step 2.

Step 1 — Add the Layer connector (admin / solo user)

  1. Open claude.ai or Claude Desktop → Customize → Connectors and click Add custom connector.

  2. Name the connection “Layer” and set the URL to:

https://mcp.app.layer.ai/mcp

  1. Click Add.

Step 2 — Connect your account (everyone)

  1. Go to Customize → Connectors → Layer → Connect.

  2. Authorize with your Layer account when prompted.

Step 3 — Verify it works

Open a new conversation and ask:

"What can I do with Layer?"

Claude should respond with Layer's capabilities. If it doesn't recognize Layer, check that the connector shows as Connected in your Connectors settings.

b. Via Claude Code (CLI)

  1. Run the following command in your terminal:

claude mcp add --transport http Layer https://mcp.app.layer.ai/mcp

  1. Launch Claude Code and run /mcp to open the MCP management panel. This shows all connected MCP servers and lets you authenticate.

  2. Select Layer and choose Authenticate. A browser window will open — log in with your Layer account.

  3. Verify by asking Claude: "What can I do with Layer?" You should see Layer's tools listed in the response.

Troubleshooting

Layer doesn't appear in my Connectors list

Your team admin needs to add the custom connector first (Step 1). If you're on a personal account, make sure you're looking under Customize → Connectors, not the integrations page.

Authorization fails or times out

Make sure you have an active Layer account at layer.ai. If you're behind a VPN or corporate firewall, the OAuth redirect may be blocked — try on a direct connection.

Claude doesn't seem to use Layer tools

Start a new conversation after connecting — existing conversations won't pick up new connectors. Also check that the connector status shows "Connected" (not just "Added").


any questions? please contact [email protected]

Did this answer your question?