Skip to main content
MCPOfficial source
Official MCP Registry

three.ws Concierge

Ask any website's AI concierge a grounded question, and generate the embed to add one to a site.

What does this entry provide?

three.ws Concierge is an MCP server entry in ChatHome Skills, sourced from Official MCP Registry and scored 93/100. This catalog entry is the indexed repository nirholas/three.ws. The command is the string recorded from the source. It is not proof of a one-click install in every client. An MCP server exposes callable tools to a client. An agent skill is an instruction package. This entry is an MCP server record. ChatHome Skills discovered this entry through its daily ingest of official registries and GitHub discovery queries, and re-verifies repository facts, freshness signals, and install metadata on every run. The quality score weighs official-registry presence, a working install command, license clarity, GitHub star trajectory, and repository freshness — the full formula is on the Sources page. Review the repository push date and repository status before adding three.ws Concierge to a production agent workflow; the snapshot facts below are point-in-time values from the most recent ingest.

Snapshot facts

  • Source: Official MCP Registry
  • Quality: 93/100
  • Install method: npm
  • Repo: nirholas/three.ws
  • Repo updated: Oct 10, 2026
  • Seen: Nov 20, 2026
Install command
npx -y @three-ws/concierge-mcp@0.1.1
Install method: npm

The command is the string recorded from the source. It is not proof of a one-click install in every client.

Review the upstream source before running this command.

Quality
93
Repo stars
0
Repo updated
Oct 10, 2026
How is this score calculated?
  • Weighted across six dimensions, scored 0-100.
  • Provenance (25): listed in an official or curated registry.
  • Installability (25): a runnable install command beats a declared transport, which beats a source checkout.
  • Maintenance (20): recency of the last push; archived entries score zero here.
  • Adoption (20): repository stars on a log curve.
  • Documentation (10) and description quality (10): parsed manifest or SKILL.md frontmatter, and a real description rather than a placeholder.
  • Dimensions a source cannot supply — registries publish no stars or license — are excluded from the entry's own average instead of counted as zero, so entries stay comparable across sources.
  • Security signals subtract up to 20. Scores are recomputed on every daily ingest.
MCP manifest
{
  "_meta": {
    "io.modelcontextprotocol.registry/official": {
      "status": "active",
      "isLatest": true,
      "updatedAt": "2026-10-10T20:46:44.423817Z",
      "publishedAt": "2026-10-10T20:46:44.423817Z",
      "statusChangedAt": "2026-10-10T20:46:44.423817Z"
    }
  },
  "server": {
    "name": "io.github.nirholas/concierge-mcp",
    "icons": [
      {
        "src": "https://three.ws/three-ws-mcp-icon.svg",
        "sizes": [
          "any"
        ],
        "mimeType": "image/svg+xml"
      }
    ],
    "title": "three.ws Concierge",
    "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
    "version": "0.1.1",
    "packages": [
      {
        "version": "0.1.1",
        "transport": {
          "type": "stdio"
        },
        "identifier": "@three-ws/concierge-mcp",
        "runtimeHint": "npx",
        "registryType": "npm",
        "environmentVariables": [
          {
            "name": "THREE_WS_BASE",
            "format": "string",
            "default": "https://three.ws",
            "description": "Base URL of the three.ws API that serves POST /api/concierge, and the origin the generated embed snippets point at. Override only when self-hosting or targeting a preview deployment."
          },
          {
            "name": "THREE_WS_TIMEOUT_MS",
            "format": "number",
            "default": "45000",
            "description": "Per-request timeout in milliseconds for the concierge answer call. The answer streams over a free-first LLM chain that can fail over across providers, so the default leaves room for one slow rung."
          },
          {
            "name": "CONCIERGE_PAGE_TIMEOUT_MS",
            "format": "number",
            "default": "12000",
            "description": "Timeout in milliseconds for fetching a page that concierge_ask grounds its answer in. Separate from the answer timeout so a slow site cannot consume the whole call."
          }
        ]
      }
    ],
    "repository": {
      "url": "https://github.com/nirholas/three.ws",
      "source": "github",
      "subfolder": "packages/concierge-mcp"
    },
    "websiteUrl": "https://three.ws/concierge",
    "description": "Ask any website's AI concierge a grounded question, and generate the embed to add one to a site."
  }
}
Report an issue with this listing