Skip to content

Arbo MCP server

com.getarbo/arbogetarbo.comstreamable-httpv1.0.1Scanned

Site structure for client redesigns. Page tree with URL, keyword, status and redirect plan export.

12 tools: 1 can delete or cancel things, 7 change things and 4 only read.

Tools: 1 delete, 0 irreversible, 7 write, 4 read.
  • Reads private data · 0
  • Reads text others write · 3
  • Can send data out · 3
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here can delete or cancel things. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Arbo MCP server can do

The Arbo MCP server is published by getarbo.com, a domain the MCP Registry verified. It has 12 tools. It can delete or cancel things (1 tool: delete_node) and create or change data (7 tools: create_project, create_node, update_node, generate_wireframe and 3 more). 4 tools only read.

3 tools bring in text other people can write (get_review, list_review_comments and get_change_list); 3 tools can send data out of your system (create_node, update_node and set_comment_resolved).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 13, 2026, version 1.0.1.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

12 tools

Deletes or cancels 1Claude Code: deny
  • delete_nodeRemove a page from the sitemap (soft delete, can be restored via version history)No findings

    Why it is in this group

    • The name "delete_node" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Remove a page from the sitemap (soft delete, can be restored via version history)
Changes things 7Claude Code: ask
  • bulk_create_nodesCreate multiple pages at once from a tree structure.No findings

    Why it is in this group

    • The name "bulk_create_nodes" says "create": a word for a change.
    Description, as the server wrote it
    Create multiple pages at once from a tree structure. Ideal for generating a full sitemap with zoning, entry points, CTAs and tags.
  • create_nodeAdd a page to a project's sitemap.Can send data outNo findings

    Why it is in this group

    • The name "create_node" says "create": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.seo.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Add a page to a project's sitemap. Use parent_id to create hierarchy. Supports rich data: zoning blocks (wireframe layout), entry points (traffic sources), CTAs, and tags.
  • create_projectCreate a new project.No findings

    Why it is in this group

    • The name "create_project" says "create": a word for a change.
    Description, as the server wrote it
    Create a new project. Returns the project ID. Add nodes with create_node after.
  • generate_wireframeGenerate a lo-fi wireframe HTML for a page using AI.No findings

    Why it is in this group

    • The name "generate_wireframe" says "generate": a word for a change.
    Description, as the server wrote it
    Generate a lo-fi wireframe HTML for a page using AI. The wireframe is saved to the node's zoningHtml field. Requires zoning blocks to be defined on the node. Costs 1 AI credit (or uses BYOK key). Returns the generated HTML.
  • list_review_commentsComments left in the review (wireframe or mockup), pinned on a version and a breakpoint.Reads text others writeNo findings

    Why it is in this group

    • The name "list_review_comments" says "list": a word for reading.
    • The description's first verb is "comment": a word for a change.

    What it can reach

    • untrusted-inputThe name "list_review_comments" says "list" and "review": it reads comments, which other people can write.
    • untrusted-inputThe description's first sentence says "comment" and "review": it reads comments, which other people can write.
    Description, as the server wrote it
    Comments left in the review (wireframe or mockup), pinned on a version and a breakpoint. Set open_only true to get only the unresolved ones.
  • set_comment_resolvedMark a review comment resolved or open again.Can send data outNo findings

    Why it is in this group

    • The name "set_comment_resolved" says "set": a word for a change.

    What it can reach

    • outboundThe name "set_comment_resolved" says "set" and "comment": it writes something other people read.
    Description, as the server wrote it
    Mark a review comment resolved or open again. Same rights as the other write tools (write:nodes, owner or editor).
  • update_nodeUpdate a page's properties.Can send data outNo findings

    Why it is in this group

    • The name "update_node" says "update": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.seo.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Update a page's properties. For the description field: use real newlines (not literal \n). Supports rich multiline content: SEO notes, copy, sections, CTAs, FAQ, maillage, etc. No length limit. You can also set zoning_html (full wireframe HTML) and annotations (contextual notes per wireframe section).
Only reads 4Show 4 read-only tools
  • get_change_listChange list from the open review comments: deduplicated and classed (text, visual, structure), out-of-scope requests with a reply and an amendment line, and contradictions with dec…Reads text others writeNo findings

    Why it is in this group

    • The name "get_change_list" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "review": it reads comments, which other people can write.
    Description, as the server wrote it
    Change list from the open review comments: deduplicated and classed (text, visual, structure), out-of-scope requests with a reply and an amendment line, and contradictions with decided points. Costs AI credits (quality model, 3 credits), paid by the token's account or the project owner. Requires write access.
  • get_projectGet a project with its full sitemap treeNo findings

    Why it is in this group

    • The name "get_project" says "get": a word for reading.
    Description, as the server wrote it
    Get a project with its full sitemap tree
  • get_reviewReview of a project: per page, the status and version of each step (wireframe, mockup), and the latest Figma link per breakpointReads text others writeNo findings

    Why it is in this group

    • The name "get_review" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe name "get_review" says "get" and "review": it reads comments, which other people can write.
    • untrusted-inputThe description's first sentence says "review": it reads comments, which other people can write.
    Description, as the server wrote it
    Review of a project: per page, the status and version of each step (wireframe, mockup), and the latest Figma link per breakpoint
  • list_projectsList all projects accessible with this tokenNo findings

    Why it is in this group

    • The name "list_projects" says "list": a word for reading.
    Description, as the server wrote it
    List all projects accessible with this token

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write3
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:39 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
8
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 8 tools that change things (1 destructive, 7 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://getarbo.com/api/mcp: tools/list returned 12 tools to an anonymous client, including delete_node (destructive), create_project (write), create_node (write), update_node (write), generate_wireframe (write) and 3 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http arbo https://getarbo.com/api/mcp --header 'Authorization: <your-authorization>'
claude mcp add --transport http arbo https://getarbo.com/api/mcp --header 'Authorization: <your-authorization>'

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__arbo__list_projects",
      "mcp__arbo__get_project",
      "mcp__arbo__get_review",
      "mcp__arbo__get_change_list"
    ],
    "ask": [
      "mcp__arbo__create_project",
      "mcp__arbo__create_node",
      "mcp__arbo__update_node",
      "mcp__arbo__generate_wireframe",
      "mcp__arbo__bulk_create_nodes",
      "mcp__arbo__list_review_comments",
      "mcp__arbo__set_comment_resolved"
    ],
    "deny": [
      "mcp__arbo__delete_node"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as arbo. Review it before you use it.

Capability tags behind 6 rules
  • mcp__arbo__create_node outbound
  • mcp__arbo__update_node outbound
  • mcp__arbo__get_review untrusted-input
  • mcp__arbo__list_review_comments untrusted-input
  • mcp__arbo__get_change_list untrusted-input
  • mcp__arbo__set_comment_resolved outbound

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.getarbo/arbo/verify?hash=2d15f33ea0fd79b2f290bb129822aec73b0edabfcc9703eaa9363ccb2914e1b6" | jq -e '.changed == false'

Pinned hash 2d15f33ea0fd, from Nov 13, 2026, 08:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.getarbo/arbo/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.getarbo/arbo
[![Anlyon Radar](https://radar.anlyon.com/badge/com.getarbo/arbo.svg)](https://radar.anlyon.com/servers/com.getarbo/arbo)

Source: MCP Registry. Synced Oct 21, 2026, 17:55 UTC.

Arbo MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Arbo MCP server safe?

No scan can promise that. What its definitions show: 12 tools, 8 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Arbo MCP server delete data?

Yes. 1 tool can delete, cancel or revoke something: delete_node.

Can the Arbo MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Arbo MCP server have?

12 tools: 1 destructive, 7 write, 4 read. delete_node, create_project, create_node, update_node, generate_wireframe, bulk_create_nodes, list_review_comments, set_comment_resolved and 4 more.

Does the Arbo MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Arbo MCP server send data to outside services?

3 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): create_node, update_node and set_comment_resolved.

Has the Arbo MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 13, 2026.