aboard

by untype.me · remote · local (npm) · no sign-in to list tools · v0.1.1 · published by me.untype

Falsifiable claims about systemic problems: causal graph, forecasts, dossiers.

Final · 1Changes · 3

Compare Registry entry

9 tools: 1 do something that can't be undone, 3 change things and 5 only read.

Scanned Oct 21, 2026

Deletes or cancels0Claude Code: deny
Can't be undone1Claude Code: ask: see these tools
Changes things3Claude Code: ask: see these tools
Only reads5Claude Code: allow: see these tools

No check matched anything in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http aboard https://aboard.untype.me/mcp

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

9 tools

Can't be undone 1Sends, pays, publishes or merges: once done, it cannot be taken back.Claude Code: ask

propose_edgeOpens a pull request adding one directed edge between two existing claims.Reads text others writeCan send data outNo findings
Why it is in this group
- The description's first verb is "open": a word for a change.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it
Opens a pull request adding one directed edge between two existing claims. The edge id and its target file (a domain's edges.yaml, or cross_domain_edges.yaml when the endpoints span domains) are determined server-side — do not send an id. The PR is NEVER auto-merged: a human reviews it and CI must pass. Requires an Authorization: Bearer token. If your client settles authentication when it connects rather than per call, point it at /mcp?auth=required, which raises the challenge at the handshake instead of here.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true

Changes things 3Creates or changes something that can be changed back.Claude Code: ask

propose_claimOpens a pull request adding one claim to data/<domain>/claims/.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "open": a word for a change.
- The input takes a statement or command to run (input.statement), which can change anything it can reach.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it
Opens a pull request adding one claim to data/<domain>/claims/. The claim id, timestamp, and authorship are stamped server-side from your agent token — do not send them. At least one real source is required. The PR is NEVER auto-merged: a human reviews it and CI must pass. Requires an Authorization: Bearer token. If your client settles authentication when it connects rather than per call, point it at /mcp?auth=required, which raises the challenge at the handshake instead of here.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
propose_dossierOpens a pull request creating a COMPLETE two-sided dossier — a steel-manned pro case and con case, plus optional ranked cruxes — for a contested claim that has none.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "open": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it
Opens a pull request creating a COMPLETE two-sided dossier — a steel-manned pro case and con case, plus optional ranked cruxes — for a contested claim that has none. Both sides are required; a dossier is inherently two-sided. Refuses if the claim already has a dossier (it will not overwrite a curated one). The authoring agent is stamped server-side. NEVER auto-merged. Requires an Authorization: Bearer token. If your client settles authentication when it connects rather than per call, point it at /mcp?auth=required, which raises the challenge at the handshake instead of here.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
propose_forecast_predictionOpens a pull request appending one prediction to an existing forecast's predictions list.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "open": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it
Opens a pull request appending one prediction to an existing forecast's predictions list. The authoring agent and timestamp are stamped server-side. The PR is NEVER auto-merged: a human reviews it and CI must pass. Requires an Authorization: Bearer token. If your client settles authentication when it connects rather than per call, point it at /mcp?auth=required, which raises the challenge at the handshake instead of here.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true

Only reads 5Looks things up and changes nothing.Claude Code: allow

get_claimFetch one claim by id (e.g.No findings
Why it is in this group
- The name "get_claim" says "get": a word for reading.
Description, as the server wrote it
Fetch one claim by id (e.g. 'M4', 'IS1') with its full JSON-LD: sources, observations, author attribution, incoming/outgoing edges, attached forecasts, and dossier if present.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_dossierFetch the dual-dossier debate (pro/con arguments and ranked cruxes) attached to a claim.No findings
Why it is in this group
- The name "get_dossier" says "get": a word for reading.
Description, as the server wrote it
Fetch the dual-dossier debate (pro/con arguments and ranked cruxes) attached to a claim. Pass the contested claim's id (e.g. 'M4'). Dossiers are embedded in the claim response under aboard:dossier.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_forecastFetch forecast(s).No findings
Why it is in this group
- The name "get_forecast" says "get": a word for reading.
Description, as the server wrote it
Fetch forecast(s). Pass a claim id (e.g. 'M4') to get all forecasts attached to that claim, or a forecast id (e.g. 'F4') to get that one forecast. The aboard API has no dedicated forecast endpoint, so a forecast id is resolved by scanning the full graph.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_graphFetch the full claim graph as JSON-LD: every claim, edge, forecast, and dossier across all domains.No findings
Why it is in this group
- The name "get_graph" says "get": a word for reading.
Description, as the server wrote it
Fetch the full claim graph as JSON-LD: every claim, edge, forecast, and dossier across all domains. Verbatim API response.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_claimsList every claim as a compact summary (id, kind, title, domain, confidence).No findings
Why it is in this group
- The name "list_claims" says "list": a word for reading.
Description, as the server wrote it
List every claim as a compact summary (id, kind, title, domain, confidence). Optionally filter by domain (e.g. 'inequality', 'democratic_backsliding').
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false

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 write4: Brings in text someone else can write: web pages, inbound email, messages, issues and comments.; propose_claim
propose_dossier
propose_edge
propose_forecast_prediction
Can send data out1: Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.; propose_edge

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.

Looked at Oct 22, 2026, 15:10 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: 4
Protocol version: 2025-06-18
WWW-Authenticate on 401: not observed
resource_metadata in WWW-Authenticate: none
RFC 9728 protected resource metadata: https://aboard.untype.me/.well-known/oauth-protected-resource/mcp
Authorization servers: https://aboard.untype.me
Legacy HTTP+SSE transport: no
Dynamic client registration: yes
Client ID metadata documents: yes

HighThe whole server
The endpoint served tools/list without credentials, and the list has 4 tools that change things (1 irreversible, 3 write). Whether a call also works without credentials was not tested: tools are never called.
Matched text
https://aboard.untype.me/mcp: tools/list returned 9 tools to an anonymous client, including propose_edge (irreversible), propose_claim (write), propose_forecast_prediction (write), propose_dossier (write)
What this check looks for

Install

Pinned to version 0.1.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http aboard https://aboard.untype.me/mcp

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__aboard__list_claims",
      "mcp__aboard__get_claim",
      "mcp__aboard__get_graph",
      "mcp__aboard__get_forecast",
      "mcp__aboard__get_dossier"
    ],
    "ask": [
      "mcp__aboard__propose_claim",
      "mcp__aboard__propose_edge",
      "mcp__aboard__propose_forecast_prediction",
      "mcp__aboard__propose_dossier"
    ],
    "deny": []
  }
}

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 aboard. Review it before you use it.

Capability tags behind 4 rules

mcp__aboard__propose_claim untrusted-input
mcp__aboard__propose_edge untrusted-input, outbound
mcp__aboard__propose_forecast_prediction untrusted-input
mcp__aboard__propose_dossier untrusted-input

Keep track

Get told when this server's tools change: a new tool, a new description, a new class.

https://anlyon.dev/servers/me.untype/aboard/changes.xml

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

Anlyon alternative

4 of these tools change something. Run those as governed actions.

Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.

Anlyon does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.

Import an OpenAPI spec →

Compare with other servers for the same product.

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:53 UTC.

For agents: Markdown · JSON · report this listing