Seam for AI agents

Build with OpenClaw. Build access, climate, and sensor workflows.

Bring Seam’s current documentation into your OpenClaw workspace. Build the application logic for access, climate, or sensor workflows, then verify the specific devices and actions you intend to support.

Illustrative sandbox workflow with OpenClaw: consult Seam documentation, prepare scheduled guest access, then confirm the sandbox result. Access runs from Friday at 3 PM until Sunday at 11 AM. The door remains locked. No live device is operated.

OpenClaw
Illustrative sandbox workflow

Set up Seam in OpenClaw.

Use OpenClaw’s MCP command on the installation that owns your agent’s connection, then test the server before starting a workflow.

01

Add Seam’s MCP server

With a current OpenClaw installation, add the endpoint over Streamable HTTP. Then open Settings → MCP and use its scoped config editor to set auth: "oauth" on mcp.servers.seam. Select Sign in and complete browser authorization on the Gateway’s Settings page.

Run on your OpenClaw installation
openclaw mcp add seam --url https://mcp.seam.co/mcp/authenticated --transport streamable-http
02

Probe the connection

After sign-in, run the live check and confirm the documentation tools and api_request are available. If your agent runs on a separate Gateway, refresh or restart that runtime after changing its configuration.

Verify the server
openclaw mcp doctor seam --probe

A saved server entry is only configuration; this probe checks that the endpoint responds.

03

Set up the project workspace

Give the agent access to your application’s project and paste this prompt. Follow /agents.md to set up a sandbox workspace and choose how your application or agent will make Seam API requests.

Seam agent setup
Paste into your OpenClaw session
Set up Seam in this project:
fetch https://seam.co/agents.md and follow its instructions.
Choose your workspace permissions.

After signing in, check that api_request appears alongside the documentation tools. Ask your agent to list your workspaces. If Seam returns a permission-consent link, open it, select the intended scope and Viewer for read-only access, then retry. Device changes require write permissions. No Seam API key belongs in this MCP configuration.

Already connected? Replace the old Seam entry instead of adding a duplicate, reconnect or restart your client, and start a new conversation to refresh its tools. For documentation and skill packs without sign-in, use https://mcp.seam.co/mcp instead and skip OAuth login. This optional anonymous connection cannot access workspace resources.

Read the MCP guide

Troubleshoot with your agent.

Investigate device and connected-account issues, Access Grants, access codes, and API or authentication errors. Seam’s MCP delivers the troubleshooting skill pack directly to your agent, with playbooks for gathering evidence and identifying what is still unknown.

Connect with Viewer access.

Use an MCP client that supports OAuth to connect to Seam’s authenticated endpoint. Sign in and approve Viewer access to the workspace you want to investigate. Refresh the connection and start a new conversation to load its tools.

Use your existing workspace. This skill is available through MCP without a separate plugin install. Replace the placeholders with the affected resource, workspace, and incident time. The investigation uses public API evidence and stays read-only.

Set up troubleshooting
Authenticated MCP endpoint
https://mcp.seam.co/mcp/authenticated
Try this prompt
Use Seam MCP to investigate why access grant <grant_id>
in workspace <workspace_id> failed at <time and timezone>.
Use the troubleshooting skill pack. Keep the investigation
read-only and explain any missing evidence.

Prompts to try.

Give OpenClaw a concrete task, ask it to check current documentation, and review the result.

Build a request-to-access workflow

Agent prompt
Design a workflow in this application that turns an approved guest request into Seam access. Use the current API guidance, bind the request to a specific person, door, and time window, and show how the application confirms readiness and handles cancellation.

Prepare a thermostat workflow

Agent prompt
Use Seam’s current documentation to build a thermostat workflow for the supported models I specify. Check the actual capabilities, validate requested settings, and track action results. Verify the application logic in a sandbox before a real-device trial.

Route sensor events through the app

Agent prompt
Build an application workflow for supported Seam sensor events. Verify webhook signatures, deduplicate deliveries, and record delivery state for the notification destination I choose. Test the handler with fixtures and report what still needs a real-device check.

Where should I add the connection?

Add it on the installation or Gateway that owns the active agent runtime. OpenClaw documents that a reload in one CLI process does not automatically reload a Gateway running elsewhere. Probe the connection where the agent will use it.

Can OpenClaw load the Seam skills?

Yes. OpenClaw supports SKILL.md directories in the agent’s workspace skills/ and .agents/skills/ folders. Copy the relevant Seam skill directories with their supporting files into the intended workspace and confirm discovery for that agent.

Next steps.

Test in a sandbox, then connect your real devices.

Find your device
Smart Locks
Access Systems
Knowledge
Lock Capabilities
Thermostats
Sensors
© 2026 Seam Labs, Inc.