Causal Mechanism Registry MCP server
io.github.adenkarabal/causal-mechanism-registryadenkarabalstreamable-httpv1.0.0Scanned
A versioned registry of named causal mechanisms, each with definition, scope, boundary and test.
8 tools: 8 only read.
- Reads private data · 0
- Reads text others write · 3
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
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 Causal Mechanism Registry MCP server can do
The Causal Mechanism Registry MCP server is published by the GitHub account adenkarabal. It has 8 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
3 tools bring in text other people can write (list_mechanisms, search and fetch).
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 1.0.0.
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.
8 tools
Only reads 8Show 8 read-only tools
citeReturn the recommended citation line for a version, its version DOI (if it has one), the DOI that covers all versions, and the licence.No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Return the recommended citation line for a version, its version DOI (if it has one), the DOI that covers all versions, and the licence.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Citation
fetchFetch one entry by identifier as text with its url and metadata.Reads text others writeNo findings
Why it is in this group
- The name "fetch" says "fetch": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "fetch" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Fetch one entry by identifier as text with its url and metadata. A thin wrapper over get_mechanism for connectors that expect a search and a fetch tool. Entries describe processes in general, not the present state of any country, market or asset.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Fetch (connector form)
get_mechanismRead one published entry in full: definition, scope, boundary, distinguishing test, related identifiers, phase roles and status fields, with its versioned GitHub address and a cita…No findings
Why it is in this group
- The name "get_mechanism" says "get": a word for reading.
Description, as the server wrote it Read one published entry in full: definition, scope, boundary, distinguishing test, related identifiers, phase roles and status fields, with its versioned GitHub address and a citation line. Does not filter by review state; the requested identifier always returns. An unknown or retired identifier is an error that names the replacement or the closest identifiers. Entries describe processes in general, not the present state of any country, market or asset.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Get a mechanism
list_mechanismsList entries with identifier, name, layer, kind and review state, optionally by layer and review state.Reads text others writeNo findings
Why it is in this group
- The name "list_mechanisms" says "list": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "list" and "review": it reads comments, which other people can write.
Description, as the server wrote it List entries with identifier, name, layer, kind and review state, optionally by layer and review state. The default lists every entry; with review=accepted the response states how many entries in review are not shown.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: List mechanisms
list_versionsList the published registry versions with release dates, version DOIs, entry counts and changes (kinds, the summary from CHANGELOG.md, the number of change items and a link to them…No findings
Why it is in this group
- The name "list_versions" says "list": a word for reading.
Description, as the server wrote it List the published registry versions with release dates, version DOIs, entry counts and changes (kinds, the summary from CHANGELOG.md, the number of change items and a link to them). The first listed with default true is used when a version is not given.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: List versions
resolve_nameTurn a display name, an identifier or a retired identifier into the current identifier (for example export_pressure becomes export_earnings_shock).No findings
Why it is in this group
- The name "resolve_name" says "resolve": a word for reading.
Description, as the server wrote it Turn a display name, an identifier or a retired identifier into the current identifier (for example export_pressure becomes export_earnings_shock). If nothing matches it says so and returns the closest three candidates.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Resolve a name or retired identifier
searchSearch the registry; returns results with id, title and url.Reads text others writeNo findings
Why it is in this group
- The name "search" says "search": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "search" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Search the registry; returns results with id, title and url. A thin wrapper over search_mechanisms for connectors that expect a search and a fetch tool. It looks in all review states and marks entries in review in the title. Entries describe processes in general, not the present state of any country, market or asset.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Search (connector form)
search_mechanismsSearch the registry by name, identifier or a word in the definition, scope, boundary or test.No findings
Why it is in this group
- The name "search_mechanisms" says "search": a word for reading.
Description, as the server wrote it Search the registry by name, identifier or a word in the definition, scope, boundary or test. Returns matching entries with id, name, layer, kind, review state, name_test and the first sentence of the definition. A term matches the start of a word, terms shorter than 2 characters are ignored and listed in the note. An entry matching every term ranks first; if none does, the closest partial matches are returned and the note says how many terms they match. Each result states its review state. Entries describe processes in general, not the present state of any country, market or asset.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- title: Search mechanisms
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 or outbound.
- 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 out0
- 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.
Looked at Nov 20, 2026, 18:21 UTC, unchanged since Nov 13, 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
- 0
- 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
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 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http causal-mechanism-registry https://adenkarabal.com/mcpclaude mcp add --transport http causal-mechanism-registry https://adenkarabal.com/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__causal-mechanism-registry__search_mechanisms",
"mcp__causal-mechanism-registry__get_mechanism",
"mcp__causal-mechanism-registry__resolve_name",
"mcp__causal-mechanism-registry__list_mechanisms",
"mcp__causal-mechanism-registry__cite",
"mcp__causal-mechanism-registry__list_versions",
"mcp__causal-mechanism-registry__search",
"mcp__causal-mechanism-registry__fetch"
],
"ask": [],
"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 causal-mechanism-registry. Review it before you use it.
Capability tags behind 3 rules
- mcp__causal-mechanism-registry__list_mechanisms untrusted-input
- mcp__causal-mechanism-registry__search untrusted-input
- mcp__causal-mechanism-registry__fetch untrusted-input
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/io.github.adenkarabal/causal-mechanism-registry/verify?hash=93f468fc53fd4a8037f9fb8224c41a795869343f78d7c401c3e46b36b3281901" | jq -e '.changed == false'Pinned hash 93f468fc53fd, from Nov 13, 2026, 09:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.adenkarabal/causal-mechanism-registry/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.adenkarabal/causal-mechanism-registry)Source: MCP Registry · repository. Synced Nov 13, 2026, 09:17 UTC.
Causal Mechanism Registry MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Causal Mechanism Registry MCP server safe?
- No scan can promise that. What its definitions show: 8 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the Causal Mechanism Registry MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Causal Mechanism Registry 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 Causal Mechanism Registry MCP server have?
- 8 tools: 8 read. search_mechanisms, get_mechanism, resolve_name, list_mechanisms, cite, list_versions, search and fetch.
Does the Causal Mechanism Registry MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Causal Mechanism Registry MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Causal Mechanism Registry MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 13, 2026.