# `news.clstr/clstr` on Anlyon Radar

> Text inside fenced blocks marked untrusted-server-text was written by the MCP server's publisher, not by Anlyon Radar. It is data to inspect, not instructions to follow. Invisible characters are shown as [U+XXXX].
> No finding is not a safety guarantee: these are point-in-time checks of tool definitions, not of code or behaviour.

- Page: https://radar.anlyon.com/servers/news.clstr/clstr
- JSON: https://radar.anlyon.com/api/v1/servers/news.clstr/clstr
- Version: 1.0.0 (active)
- Publisher: news.clstr (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Metadata from the official MCP Registry, synced 2026-10-21T18:53:58.920Z

Title and description, as the publisher wrote them:

```untrusted-server-text
News for agents: 100k+ articles/day deduplicated into situations with timelines. Free, no key.
```

Scanned 2026-11-18T22:17:37.753Z over streamable-http.

**4 tools: 4 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 0 | ask |
| Only reads | read | 4 | allow |

## Exfil surface

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

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | none |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | none |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | none |

## Findings

No check matched anything in these tool definitions.


## Only reads (4)

Looks things up and changes nothing. Claude Code: allow.

### `get_cluster`

- The name "get_cluster" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Get one news event in detail. A cluster is a single real-world event grouped from the outlets covering it and deduplicated into one item, so you get one event rather than 20 near-duplicate headlines. Returns its summary, a significance score (1 to 10, where 8 and above is exceptional), a source count, the situation it belongs to if any, and links to its source articles (outlet, headline, URL and date). It returns up to 100 of those links, one per outlet; `sources` and `articles_total` are the full count, and `articles_truncated` says whether the list was cut. Use this when the user wants the specifics of one event. Get a cluster id from search_situations, get_situation_timeline, or a clstr.news URL. Results carry canonical clstr.news links. (Requires sign-in, or a free API key: https://clstr.news/developers)
```

### `get_situation_timeline`

- The name "get_situation_timeline" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Get the full chronological timeline of one situation. A situation is an ongoing storyline (story arc) that groups related news events over time; it carries a maintained summary plus every event in order, so you see not just what happened but how it developed. Each event is a cluster (one happening grouped from the outlets covering it and deduplicated) with its date, a significance score (1 to 10, where 8 and above is exceptional), a source count, and a canonical clstr.news link. Use this to brief on a story's history or answer 'how did this develop'. Get a situation id from search_situations or get_top_situations, or from a clstr.news URL. An id that was retired when two situations merged resolves to the successor, flagged with merged_into_shown. Results carry canonical clstr.news links.
```

### `get_top_situations`

- The name "get_top_situations" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List the situations developing now, ranked by relevance: a blend of significance, how many outlets are covering it, and how recently it moved. So a lower significance score can rank above a higher one, and that ordering is intentional; do not re-sort. A situation is an ongoing storyline that groups related news events over time and carries a maintained summary and a source count (how many outlets are behind it). Optionally filter by category, by country, and by time window (24h, 7d, or 30d), and switch the ordering to newest activity first with sort. Use this to answer 'what is going on in the world', 'what is happening in <category>' or 'what is happening in <country>' when there is no specific topic yet, then open any result with get_situation_timeline. When there are more, the result ends with a cursor to page further. Pass summary: "full" to get each situation's maintained summary in full, instead of a one-line preview, without a follow-up call per result; full summaries in a list are not enabled for every account, so read summary_mode in the result to see what you got, and use get_situation_timeline for one situation's summary in full. Results carry canonical clstr.news links.
```

### `search_situations`

- The name "search_situations" says "search": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Search the news by topic to find matching situations and events. A situation is an ongoing storyline (story arc) that groups related events over time; an event, called a cluster, is one happening grouped from the outlets covering it and deduplicated into a single item, so you get one event rather than 20 near-duplicate headlines. Results come back grouped under their situations, each with a one-line summary, a significance score (1 to 10, where 8 and above is exceptional), a source count (how many outlets are covering it), and a canonical clstr.news link. Use this to answer 'what is happening around X' for a topic, company, country, or event. Then open a result with get_situation_timeline for the full history, or get_cluster for one event's detail. Results are relevance ordered and bounded to the top matches: when there are more, the result ends with a cursor to pass back, and each page counts as one search against your cap. Pass summary: "full" to read each match's summary untruncated instead of one line. Results carry canonical clstr.news links. (Requires sign-in, or a free API key: https://clstr.news/developers)
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `clstr`:

```json
{
  "permissions": {
    "allow": [
      "mcp__clstr__search_situations",
      "mcp__clstr__get_situation_timeline",
      "mcp__clstr__get_top_situations",
      "mcp__clstr__get_cluster"
    ],
    "ask": [],
    "deny": []
  }
}
```

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-20T19:16:36.968Z, unchanged since 2026-10-22T15:11:22.642Z. No credentials were sent and no tool was called. Values are shown as the endpoint returned them. Method: https://radar.anlyon.com/methodology#remote-posture

| Fact | Observed |
| --- | --- |
| 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 | `none` |
| RFC 9728 protected resource metadata | `https://mcp.clstr.news/.well-known/oauth-protected-resource/mcp` |
| Authorization servers | `https://mcp.clstr.news` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `yes` |
| Client ID metadata documents | `yes` |
