# Bezlo MCP (npm · bezlo-mcp)

Bezlo MCP: device-preset mobile screenshots, layout diagnostics, see-and-act sessions.

- Trust score: 69/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- npm · `bezlo-mcp`: 69/100 (this document), [markdown](https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp.md), [page](https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp)

## Channel facts

- Registry: `npm`
- Package: `bezlo-mcp`
- Version: `0.1.2`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Supply Chain Security**: 100/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 0 of 1 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 7 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 64/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1165 tokens (~89/item across 13 items; 13 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 81/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 44% of tool parameters carry a description.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "session_route_clear" implies "remove" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 13 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the Bezlo MCP server?

Bezlo MCP runs locally as an npm package, launched with npx -y bezlo-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add gba3124-bezlo-mcp -- npx -y bezlo-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "gba3124-bezlo-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "bezlo-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "gba3124-bezlo-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "bezlo-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add gba3124-bezlo-mcp -- npx -y bezlo-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "gba3124-bezlo-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "bezlo-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add gba3124-bezlo-mcp --command npx --arg -y --arg bezlo-mcp
```

### Hermes

```yaml
mcp_servers:
  gba3124-bezlo-mcp:
    command: "npx"
    args: ["-y", "bezlo-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "gba3124-bezlo-mcp": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "bezlo-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add gba3124-bezlo-mcp -t stdio -c npx -a -y bezlo-mcp
```

### Other

```json
{
  "mcpServers": {
    "gba3124-bezlo-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "bezlo-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 69, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-31 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 66)

First indexed and scored.

## MCP tools (13)

### `list_sessions` (~19 tokens)

List open interactive sessions (id, url, device).

### `list_devices` (~49 tokens)

List the Bezlo device presets (iPhone, Android, iPad, desktop) with width/height/dpr. Pass an optional category to filter.

Input parameters:

- `category` (string): Optional category filter

### `capture_page` (~206 tokens)

Render a URL in Chrome for Testing with a real device preset (viewport, DPR, mobile UA/touch) and return a PNG screenshot plus layout diagnostics (horizontal overflow + offending elements, safe-area insets, media features, viewport/screen sizes). Use it to check mobile/web layouts, responsive bugs, and before/after fixes.

Input parameters:

- `darkMode` (boolean): Render with prefers-color-scheme: dark
- `device` (string): Device id or name, e.g. iphone16, "iPhone 15 Pro", ipadpro11, px10 (default iphone16)
- `fullPage` (boolean): Capture the full scrollable page (default: viewport only)
- `scrollY` (number): Scroll to this Y offset before capture
- `selector` (string): Wait for this CSS selector before capture
- `url` (string, required): http(s) URL to render
- `waitMs` (number): Extra wait after load before capture (default 1200)

### `capture_matrix` (~109 tokens)

Capture the same URL on several device presets in one call (matrix). Returns each device screenshot + layout; use it to compare iPhone/iPad/Android side by side.

Input parameters:

- `devices` (array): Device ids/names (1–6), e.g. ["iphone16","ipadpro11","px10"]
- `engine` (string)
- `fullPage` (boolean)
- `selector` (string)
- `url` (string, required)
- `waitMs` (number)

### `session_assert` (~73 tokens)

Assert checks against an open session and return pass/fail per check. Checks: visible, text (contains), inputValue (equals), urlIncludes, noOverflow, noConsoleErrors, elementCount (min). Use after actions to verify a flow.

Input parameters:

- `checks` (array, required)
- `sessionId` (string, required)

### `run_flow` (~179 tokens)

Open a session, execute a list of actions in one call, and return a compact step report plus the final screenshot (optionally a screenshot per step). Use it to run a whole test flow: sign in, open a drawer, submit a form — then inspect the report.

Input parameters:

- `asserts` (array): Assertions evaluated after the last step (same schema as session_assert checks)
- `device` (string)
- `engine` (string): Chromium (default) records video; webkit is a Safari-flavoured approximation
- `record` (boolean): Record a webm video of the flow (and a gif when ffmpeg is available)
- `screenshots` (string): Save final only (default), every step, or none
- `steps` (array, required): Action objects, same schema as session_act action
- `url` (string, required)

### `session_route` (~93 tokens)

Mock a network response for an open session (Playwright route). Give a URL pattern/glob and {status, json|body|contentType|headers} or {abort:true}. Use it to test error states without a real backend.

Input parameters:

- `pattern` (string, required): URL glob, e.g. **/api/items or **/missing.png
- `response` (object)
- `sessionId` (string, required)

### `session_route_clear` (~25 tokens)

Remove all network mocks from a session.

Input parameters:

- `sessionId` (string, required)

### `session_start` (~161 tokens)

Open an interactive device session on a URL (iPhone/Android/iPad preset). Returns the first observation: screenshot, labeled interactive elements (refs e1..eN with selector/role/name/rect), layout diagnostics, and console/page errors. Use session_act to tap/type/scroll, then read the fresh observation it returns.

Input parameters:

- `darkMode` (boolean)
- `device` (string): Device id or name (default iphone16)
- `engine` (string): webkit gives a Safari-flavoured approximation (no recording/CDP a11y)
- `selector` (string): Wait for this selector before observing
- `url` (string, required)
- `waitMs` (number): Wait after load before the first observation (default 800)

### `session_observe` (~47 tokens)

Re-observe an open session: fresh screenshot, interactive element list, layout diagnostics, scroll position, and any console/page errors since the last call.

Input parameters:

- `sessionId` (string, required)

### `session_act` (~100 tokens)

Perform one action in a session and return the new observation (screenshot + elements + errors). Actions: click (index | text | selector | role+name | x+y), type (target + value + submit?), scroll (dy?), press (key), goto (url), back, reload, wait (ms). Prefer the `index` of an element from the latest observation.

Input parameters:

- `action` (object, required)
- `sessionId` (string, required)

### `session_end` (~24 tokens)

Close an interactive session and its browser.

Input parameters:

- `sessionId` (string, required)

### `inspect_page` (~80 tokens)

Same device-preset render as capture_page but returns only layout diagnostics (no image) — cheaper for quick overflow/safe-area checks.

Input parameters:

- `darkMode` (boolean)
- `device` (string)
- `scrollY` (number)
- `selector` (string)
- `url` (string, required)
- `waitMs` (number)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp#diagnostics

## Score history

- 2026-11-05: 69
- 2026-11-04: 69
- 2026-11-03: 68
- 2026-11-02: 68
- 2026-11-01: 67
- 2026-10-31: 67
- 2026-10-30: 66
- 2026-10-29: 66
- 2026-10-28: 66

## Common questions

### What is the Bezlo MCP server?

Bezlo MCP is listed in the public MCP registry as io.github.gba3124/bezlo-mcp. Bezlo MCP: device-preset mobile screenshots, layout diagnostics, see-and-act sessions. This page covers its npm package (bezlo-mcp).

### Is the Bezlo MCP server safe to use?

Bezlo MCP scores 69 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Bezlo MCP server expose?

Bezlo MCP exposes 13 tools: list_sessions, list_devices, capture_page, capture_matrix, session_assert, and 8 more. Their descriptions and schemas cost roughly 1,165 tokens of context every time the server is loaded.

### Is the Bezlo MCP server still maintained?

Bezlo MCP is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Bezlo MCP server under?

Bezlo MCP declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/bezlo-mcp
- Socket report: https://socket.dev/npm/package/bezlo-mcp
- Repository: https://github.com/gba3124/bezlo
- Changelog RSS feed: https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp.json
- HTML version of this page: https://verifymcp.io/servers/gba3124-bezlo-mcp/bezlo-mcp