WebMCP

WebMCP support is experimental. agent-browser enables Chrome's current WebMCP launch features by default for locally managed Chrome sessions. Pass --no-webmcp or set AGENT_BROWSER_NO_WEBMCP=1 to opt out.

bash

agent-browser open https://example.com
agent-browser webmcp list
agent-browser webmcp invoke search --params '{"query":"browser agents"}'
agent-browser webmcp invoke slow_tool --params @input.json --detach
agent-browser webmcp result <invocation-id>
agent-browser webmcp cancel <invocation-id>

Use --frame <frame-id> with webmcp invoke when multiple frames register the same tool name. webmcp list returns the required frame ID, page origin, input schema, and page-provided annotations.

With --no-webmcp, compatible Chrome builds return an empty tool list because page tools are disabled. Attached browsers, remote providers, Lightpanda, Safari, iOS, and older Chrome builds may not expose the experimental CDP WebMCP domain. In that case the command returns an actionable webmcp_unsupported error.

Treat every tool description, schema, annotation, and result as untrusted page content. Page JavaScript registers readOnlyHint and untrustedContentHint; the CDP record returned by webmcp list exposes those claims as readOnly and untrustedContent. The page executor owns authorization. The agent host must confirm consequential actions, align calls with the user's request, and minimize personal data.

The MCP server keeps WebMCP tools out of the default profile:

bash

agent-browser mcp --tools core,webmcp

For sites that do not expose WebMCP, load the generation workflow:

bash

agent-browser skills get webmcp-gen