# `com.evalgate/evalgate` 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/com.evalgate/evalgate
- JSON: https://radar.anlyon.com/api/v1/servers/com.evalgate/evalgate
- Version: 1.0.0 (active)
- Publisher: com.evalgate (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Metadata from the official MCP Registry, synced 2026-11-05T03:17:44.780Z

Title and description, as the publisher wrote them:

```untrusted-server-text
EvalGate
Run AI evaluations, record agent traces and gate releases on evidence, with scoped org access.
```

Scanned 2026-11-18T22:17:59.931Z over streamable-http.

**17 tools: 4 change things and 13 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 4 | ask |
| Only reads | read | 13 | allow |

## Exfil surface

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

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | `mcp__evalgate__docs_search` |
| 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

- **warn** Annotations that disagree with the tool in `runs_list` ([annotation-conflict](https://radar.anlyon.com/methodology#annotation-conflict))

```untrusted-server-text
The server marks "runs_list" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "runs_list" says "run": a word for a change.
```

- **warn** Annotations that disagree with the tool in `run_explain` ([annotation-conflict](https://radar.anlyon.com/methodology#annotation-conflict))

```untrusted-server-text
The server marks "run_explain" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "run_explain" says "run": a word for a change.
```

- **info** The word "credentials" in `connector_health` ([credentials-mention](https://radar.anlyon.com/methodology#credentials-mention))

```untrusted-server-text
The description mentions "credential".

description: "..., score and timestamp; it is not a live credential or provider check."
```

- **info** The word "credentials" in `connections_graph` ([credentials-mention](https://radar.anlyon.com/methodology#credentials-mention))

```untrusted-server-text
The input schema at properties.id.description mentions "credential".

inputSchema.properties.id.description: "...returned by connections.backfill, not a credential ID."
```


## Changes things (4)

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

### `connections_roots`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

Description, as the server wrote it:

```untrusted-server-text
Page through the configured account roots (top-level connected accounts) using current administrator authority. Use it to audit which accounts are configured. Returns root metadata and a continuation cursor; never secrets, and no claim that a connector works.
```

### `project_plan`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

Description, as the server wrote it:

```untrusted-server-text
Plan the next steps for connecting a repository, reviewing an evaluation, or ingesting traces. Use it when deciding what to set up next. Returns an ordered advisory step list and a web link; it never runs, writes or accepts anything.
```

### `run_explain`

- The name "run_explain" says "run": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Findings: warn annotation-conflict

Description, as the server wrote it:

```untrusted-server-text
Explain why one run passed or failed. Use it after runs_list when you need the failing cases and where the evidence came from. Returns the run's status, its failed cases with result metadata, an evidence-lineage summary, or not_found.
```

### `runs_list`

- The name "runs_list" says "run": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Findings: warn annotation-conflict

Description, as the server wrote it:

```untrusted-server-text
List the runs of one evaluation, newest first, optionally filtered by status. Use it to find a runId or check recent results and regression signals. Returns one page of run summaries and a nextCursor for the following page.
```

## Only reads (13)

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

### `baseline_lineage`

- The description's first verb is "show": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Show which run is the published baseline for an evaluation and where its evidence came from. Use it before comparing a new run against the baseline. Returns the baseline run and version, a lineage summary, or not_found.
```

### `connections_catalog`

- The description's first verb is "inspect": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Inspect every connector definition and its exact installed setup path. Use it to choose a connector or find how to install one. Returns connector definitions with setup paths; blueprints that are unavailable and live verification are reported as separate facts.
```

### `connections_graph`

- The description's first verb is "read": a word for reading.
- Findings: info credentials-mention

Description, as the server wrote it:

```untrusted-server-text
Read how connections link to resources: canonical connection references and the exact resource revisions currently visible. Use it to trace which source revision a connection points at. Returns references and revisions; it grants no source access and runs nothing.
```

### `connections_list`

- The name "connections_list" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List the workspace's current connections with their source restrictions and freshness. Use it to see what is already connected before planning a new one. Returns connection metadata and a continuation cursor; never secrets, and it grants no source access.
```

### `connector_health`

- The description's first verb is "show": a word for reading.
- Findings: info credentials-mention

Description, as the server wrote it:

```untrusted-server-text
Show the latest persisted connector-readiness reports, optionally for one connector. Use it to check whether an integration was set up and scored. Returns each report's status, score and timestamp; it is not a live credential or provider check.
```

### `docs_search`

- The name "docs_search" says "search": a word for reading.
- Tag private-data: The name "docs_search" says "search" and "doc": it reads documents.

Description, as the server wrote it:

```untrusted-server-text
Search EvalGate's public REST API operations by keyword. Use it to find the right endpoint before writing an API call. Returns matching operations with method, path, summary and operationId; for product guides use the documentation MCP instead.
```

### `evaluations_list`

- The name "evaluations_list" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List the organization's evaluations, optionally filtered by lifecycle status. Use it to find an evaluationId before calling runs_list or baseline_lineage. Returns one page of evaluation summaries and a nextCursor for the following page.
```

### `project_inspect`

- The name "project_inspect" says "inspect": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Snapshot the evaluations and project keys in the authenticated organization. Use it first to orient in an unfamiliar workspace or to resolve a projectKey. Returns evaluation summaries, a bounded count and a web link; changes nothing.
```

### `quality_check_get`

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

Description, as the server wrote it:

```untrusted-server-text
Get one Portable Quality execution owned by the authenticated principal.
```

### `quality_contexts_list`

- The name "quality_contexts_list" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List Portable Quality work contexts visible to the authenticated principal.
```

### `quality_plans_list`

- The name "quality_plans_list" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List frozen Portable Quality plans visible to the authenticated principal.
```

### `quality_resources_list`

- The name "quality_resources_list" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
List the immutable resource revisions currently attached to a visible Portable Quality context.
```

### `quality_standards_resolve`

- The name "quality_standards_resolve" says "resolve": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Resolve approved Portable Quality standards for the authenticated principal and exact task scope.
```

## Claude Code permissions

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

```json
{
  "permissions": {
    "allow": [
      "mcp__evalgate__project_inspect",
      "mcp__evalgate__evaluations_list",
      "mcp__evalgate__baseline_lineage",
      "mcp__evalgate__docs_search",
      "mcp__evalgate__connector_health",
      "mcp__evalgate__connections_list",
      "mcp__evalgate__connections_catalog",
      "mcp__evalgate__connections_graph",
      "mcp__evalgate__quality_contexts_list",
      "mcp__evalgate__quality_resources_list",
      "mcp__evalgate__quality_plans_list",
      "mcp__evalgate__quality_check_get",
      "mcp__evalgate__quality_standards_resolve"
    ],
    "ask": [
      "mcp__evalgate__project_plan",
      "mcp__evalgate__runs_list",
      "mcp__evalgate__run_explain",
      "mcp__evalgate__connections_roots"
    ],
    "deny": []
  }
}
```

Capability tags behind the rules (a read tagged outbound is asked about):

- `mcp__evalgate__docs_search`: private-data

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-19T17:22:45.227Z, unchanged since 2026-11-12T10:21:00.863Z. 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 | `4` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `https://www.evalgate.com/.well-known/oauth-protected-resource/api/mcp` |
| Authorization servers | `https://www.evalgate.com` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `yes` |
| Client ID metadata documents | `no` |

- **warn** Tools that change things, listed to anyone ([unauth-remote-writes](https://radar.anlyon.com/methodology#unauth-remote-writes))

```untrusted-server-text
The endpoint served tools/list without credentials, and the list has 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.

https://www.evalgate.com/api/mcp: tools/list returned 17 tools to an anonymous client, including project_plan (write), runs_list (write), run_explain (write), connections_roots (write)
```

Posture changes:

- 2026-11-12T10:21:00.863Z: Tools that change things, listed without credentials `3` -> `4`
