Mmcp.market

MCP server

by tuskcms.com·com.tuskcms/mcp·v0.3.1

Client editing for sites you built: register, scan, publish, go live, invite the client.

C58/100grade C
What users say
No reviews yet
Be the first
Safety scan
C58/100

full report

Adoption
New

Little public usage data yet

Reviews

Write one

Nobody has reviewed MCP yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

MCP tools (18, 5 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • create_sitewrite action

    Create a new site in this studio. `name` is required; `domain` (the live hostname) and `previewUrl` (an https address Tusk can reach to scan and preview) are strongly recommended so the site is connectable straight away. `deployHook` (an https deploy hook, or a vercel://token@team/project URL) lets a publish rebuild the host. Returns the new site and its slug. May return a 402 if the studio is ove

  • create_tokenwrite action

    Mint a new personal access token for this studio and return it once. Use it to provision another headless client (a CI job, a second machine). The plaintext is shown only in this result; store it immediately. Pass `site` to bind the new token to a single site (least privilege — it can act on that site and no other) and `scopes` to record the capabilities it needs. Omit both for a studio-wide token

  • deploy_statuswrite action

    Check where the host's rebuild stands after a publish (queued/building/ready/error). Pass the deployment `id` from a publish result when you have it; otherwise the newest production deployment is checked. Only this tool ever reports a deploy as actually live.

  • get_connect_guide

    Fetch the public Tusk integration guide (the data-tusk marking grammar, how scanning works, how to pull and apply a snapshot). Read this once before marking up or scanning a site so you use the right conventions. Needs no token.

  • get_instruction

    A ready-to-follow instruction, tailored to THIS site, for wiring its build to Tusk (the build token, the pull command, how to apply the snapshot). Use get_connect_guide instead for the general grammar before a site exists.

  • get_schema

    The site's editable schema: every page and the fields on it (id, label, kind, and the CSS bind that locates each field in the page HTML). This is what the editor edits and what `@tuskcms/sdk` applies at build time. Read it after a scan to see what became editable.

  • get_setup

    The connection checklist for a site: is it reachable, does it have editable fields, a deploy hook, a client seat, and has it ever been published — plus the next step to do. Use it to decide what to do next when connecting a site.

  • get_site

    Full detail for one site: its pages (with editable fields), clients, project record, files, invoices and recent events. Use it to inspect a site before scanning or publishing.

  • invite_client

    Invite the site owner (the client) to edit one site. Tusk emails them a link; they choose a password and land in the editor for that site only, seeing just the fields marked editable. Call it last, after publish and deploy_status confirm the site goes live. Needs the `invite` capability on a scoped token. Never invite an address the developer has not given you.

  • list_sites

    List every site in this studio with its slug, domain, preview URL, page count, clients, and last publish/deploy status. Use the returned `slug` as the identifier for every other site tool. (Backed by the studio overview, which also reports the studio plan and site allowance.)

  • list_tokens

    List this studio's Tusk access tokens (label, masked preview, last used, created). Secrets are never returned. Use it to audit which tokens exist before revoking one.

  • publishwrite action

    Publish a site: flip its pages' draft content to live, write the snapshot the build reads, and (in hook mode) ping the deploy hook to rebuild the host. Omit `pages` to publish every page with unpublished changes, or pass specific page keys. `note` is an optional short publish note. Read the returned `status` (published, deploy_requested, deploy_failed, no_hook, feed) then poll deploy_status for th

  • revoke_token

    Revoke one of this studio's access tokens by its id (from list_tokens). Revocation takes effect immediately. Do not revoke the token this server is currently using unless you mean to disconnect it.

  • rotate_build_token
  • save_fields

    Set the editable fields for a site by hand, or keep a chosen subset after a scan. `pages` is an array of { key, title?, url?, remove?, fields:[{ id, label, kind, bind?, auto? }] }; kinds are text, textarea, html, photos, file, files, list. Prefer scan_site to discover fields; use this to curate or define them precisely.

  • scan_site

    Scan one or more live page URLs to discover editable content and write it into the site's schema. Pass the page addresses in `urls`. `auto` controls whether unmarked elements are read: leave it unset to read unmarked elements only on pages with no data-tusk marks at all, true to always read unmarked elements, false to only honour explicit data-tusk marks. Returns the fields found per page and what

  • set_deploywrite action

    Record how a publish reaches the live site, after YOU have wired it in the developer's OWN host. Tusk never touches the host itself and never stores host credentials: it only remembers the deploy hook URL so a future client Publish can trigger a rebuild. Typical flow for a git-linked host: in the host (e.g. Cloudflare Pages → Settings → Deploy Hooks) create a deploy hook, ensure the build runs `tu

  • whoami

    Return the studio user this access token belongs to, and the sites it can see. Call this first to confirm the token works and which studio you are acting as.

Public scan report

scanner v0.1.5 · 2026-09-19 · same rubric, same numbers if you re-run it

no findings
  • Code scan8 source files scanned25/25
  • Live reliabilityno gateway calls yet and no remote to proben/a
  • Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancerepository not readable: repo not found3/15
  • Maintainer identityverified namespace with website, no repo4/10
Overall 58/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

Install directly

Runs npx -y @tuskcms/mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add mcp -- npx -y @tuskcms/mcp
Add to Cursor

MCP: common questions

Is MCP server safe?
With care: it is graded C, so read the findings first (58/100). Read the MCP safety report
How do I install MCP?
It runs on your machine. Copy the Claude Code, Claude Desktop or Cursor config from the install section.
Does MCP need an API key?
Yes. The registry entry asks for TUSK_PAT.
Is MCP maintained?
The latest release is v0.3.1.
What can I use instead of MCP?
Servers from other publishers that do the same job: Katla MCP server, Malinois MCP server and DNS Doctor MCP server. Compare all MCP alternatives.

Alternatives to MCP

Same job from other publishers: the closest match first, then the best rated.

All MCP alternatives →
  • Katla
    GDPR/CCPA cookie consent: scan sites for cookies, track consent rates, generate privacy policies.
    B
  • Malinois
    Check a live app you own for public databases, leaked keys and exposed files.
    C
  • DNS Doctor
    Scan, fix, verify and monitor DNS: SPF, DMARC, DKIM, propagation, health, expiry. Validated fixes.
    A
  • OpenQR
    Generate, edit and track dynamic (editable) QR codes with scan analytics. Hosted MCP and REST API.
    A
  • Benspdf
    Read PDFs locally: text, pages, metadata, scan detection, layout, permissions, render, OCR.
    A

More from tuskcms.com