webmcp-lab.vercel.app
Lighthouse for WebMCP: test whether an agent can actually use your site reliably.
First seen 2026-11-05 · Last seen 2026-11-05 · Source: webmcp.com
Answer 2 · Act 3 · Transact 0 · Score: Not yet scored
Tool inspector
run_eval
Run one task fixture against one target site through that site's WebMCP tools, then check the site's own state to decide whether the task really happened. Returns a verdict and a run id.
Input schema
| Property | Type | Required | Description |
|---|---|---|---|
targetId |
string | yes | Which site to evaluate |
fixtureId |
string | yes | Which task to run |
Raw JSON schema
{
"type": "object",
"properties": {
"targetId": {
"type": "string",
"enum": [
"field-notes",
"field-notes-v2",
"field-notes-fixed"
],
"description": "Which site to evaluate"
},
"fixtureId": {
"type": "string",
"enum": [
"save-neurodivergent-posts"
],
"description": "Which task to run"
}
},
"required": [
"targetId",
"fixtureId"
]
}
Similar websites
morphus.myles4miles.chatgpt.site
Morphus is an AI-native workspace for designing worlds and generating playable game experiences.
damnpdfs.com
Install the DamnPDFs Codex plugin, pair a browser workspace and use PDF, image and file tools from your coding agent. Includes setup and privacy limits.
insignifi4nt.github.io
broCAD/M: local-first CAD and CAM for parametric design, wire EDM planning, simulation and controller export.
cabinet.varera.studio
The Cartographer's Cabinet