# io.github.jeet-dhandha/agent-webbridge (npm · agent-webbridge)

Drive your real, logged-in Chrome (many profiles, parallel tabs) from any MCP client. Local only.

- Trust score: 78/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- npm · `agent-webbridge`: 78/100 (this document), [markdown](https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge.md), [page](https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge)

## Channel facts

- Registry: `npm`
- Package: `agent-webbridge`
- Version: `1.3.3`
- 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 25 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 63/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 2586 tokens (~161/item across 16 items; 16 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 77/100
  - Stability observed for 23 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 90/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 70% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 16 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 17 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 io.github.jeet-dhandha/agent-webbridge MCP server?

io.github.jeet-dhandha/agent-webbridge runs locally as an npm package, launched with npx -y agent-webbridge. 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 jeet-dhandha-agent-webbridge -- npx -y agent-webbridge
```

### Cursor

```json
{
  "mcpServers": {
    "jeet-dhandha-agent-webbridge": {
      "command": "npx",
      "args": [
        "-y",
        "agent-webbridge"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "jeet-dhandha-agent-webbridge": {
      "command": "npx",
      "args": [
        "-y",
        "agent-webbridge"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add jeet-dhandha-agent-webbridge -- npx -y agent-webbridge
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add jeet-dhandha-agent-webbridge --command npx --arg -y --arg agent-webbridge
```

### Hermes

```yaml
mcp_servers:
  jeet-dhandha-agent-webbridge:
    command: "npx"
    args: ["-y", "agent-webbridge"]
```

### Netclaw

```json
{
  "McpServers": {
    "jeet-dhandha-agent-webbridge": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "agent-webbridge"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add jeet-dhandha-agent-webbridge -t stdio -c npx -a -y agent-webbridge
```

### Other

```json
{
  "mcpServers": {
    "jeet-dhandha-agent-webbridge": {
      "command": "npx",
      "args": [
        "-y",
        "agent-webbridge"
      ]
    }
  }
}
```

## 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 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-30 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 53 to 57. That category is still filling its 30-day observation window: 16 days of observed history at the previous scan, 17 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 75, +1)

No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 74, +1)

No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-22 (score 72, +4)

- [functional improvement] Stability: unverified → 0.30

### 2026-10-13 (score 68, +29)

- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional] First check of Schema quality: fail
- [functional] First check of Tool coverage: 70
- [functional] First check of Destructive annotations: pass
- [functional] First check of Schema quality: fail
- [functional] First check of Schema quality: good

## MCP tools (16)

### `browser_status` (~30 tokens)

Fleet health: which Chrome profiles have a daemon up and the extension connected. Call first if other tools fail.

### `browser_navigate` (~243 tokens)

Open a URL in the user's real, logged-in Chrome. Returns {url, tabId}.

Input parameters:

- `active` (boolean): false opens/loads the tab in the background without taking focus (use for parallel tabs). Default: focus the tab.
- `group_title` (string): Label for the session's tab group. Pass the same value on every navigate of a session (each call re-applies it).
- `newTab` (boolean): Open in a new tab instead of reusing the current one.
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.
- `url` (string, required)

### `browser_snapshot` (~155 tokens)

Accessibility tree of the page with @e refs for interactive elements. Prefer this over screenshots for reading a page.

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_click` (~155 tokens)

Click an element by @e ref or CSS selector (synthetic click).

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `selector` (string, required)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_trusted_click` (~191 tokens)

Click with a real mouse event (isTrusted=true). Use when click is ignored by the page.

Input parameters:

- `button` (string)
- `clickCount` (number)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `selector` (string)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.
- `x` (number)
- `y` (number)

### `browser_fill` (~164 tokens)

Type into an input, textarea or contenteditable by @e ref or CSS selector.

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `selector` (string, required)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.
- `value` (string, required)

### `browser_evaluate` (~164 tokens)

Run JavaScript in the page and return the result (async/await supported). Wrap code in an IIFE.

Input parameters:

- `code` (string, required)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_screenshot` (~182 tokens)

Screenshot the page or one element. Saves to `path` and returns the file path, not base64.

Input parameters:

- `format` (string)
- `path` (string): Absolute file path to write.
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `selector` (string)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_list_tabs` (~144 tokens)

List open tabs with url, title and group.

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_find_tab` (~160 tokens)

Make an already-open tab current, by its full URL.

Input parameters:

- `active` (boolean)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.
- `url` (string, required)

### `browser_close_tab` (~138 tokens)

Close one tab.

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_close_session` (~149 tokens)

Close every tab opened by a session. Call when a task is done.

Input parameters:

- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_upload` (~154 tokens)

Set files on a file input.

Input parameters:

- `files` (array, required)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `selector` (string, required)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_file_chooser_intercept` (~177 tokens)

Toggle interception of native OS file chooser dialogs. Suppresses Finder/Explorer dialogs during programmatic uploads.

Input parameters:

- `enabled` (boolean, required): true to intercept and suppress native dialogs, false to restore normal behavior.
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_network` (~169 tokens)

Capture network requests: cmd = start | stop | list | detail.

Input parameters:

- `cmd` (string, required)
- `filter` (string)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `requestId` (string)
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

### `browser_save_as_pdf` (~158 tokens)

Save the current page as a PDF.

Input parameters:

- `paper_format` (string)
- `path` (string)
- `profile` (string): Chrome profile name, email or directory (e.g. "Work"). Omit to use the default profile.
- `session` (string): Task id that groups tabs into one Chrome tab group (default "mcp", which every call that omits it shares). Use ONE session per task and the same value for every call of that task, including parallel…
- `tabId` (number): Target a specific tab (the tabId returned by navigate, or list_tabs). REQUIRED for parallel work: without it a call goes to the profile's shared current tab and queues behind other calls.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge#diagnostics

## Score history

- 2026-11-05: 78
- 2026-11-04: 78
- 2026-11-03: 77
- 2026-11-02: 77
- 2026-11-01: 76
- 2026-10-31: 76
- 2026-10-30: 76
- 2026-10-29: 75
- 2026-10-28: 75
- 2026-10-27: 74
- 2026-10-26: 74
- 2026-10-25: 73
- 2026-10-24: 73
- 2026-10-23: 72
- 2026-10-22: 72
- 2026-10-21: 68
- 2026-10-20: 68
- 2026-10-19: 68
- 2026-10-18: 68
- 2026-10-17: 68
- 2026-10-16: 68
- 2026-10-15: 68
- 2026-10-14: 68
- 2026-10-13: 68
- 2026-10-12: 39
- 2026-10-11: 39

## Common questions

### What is the io.github.jeet-dhandha/agent-webbridge MCP server?

io.github.jeet-dhandha/agent-webbridge is an MCP server listed in the public MCP registry as io.github.jeet-dhandha/agent-webbridge. Drive your real, logged-in Chrome (many profiles, parallel tabs) from any MCP client. Local only. This page covers its npm package (agent-webbridge).

### Is the io.github.jeet-dhandha/agent-webbridge MCP server safe to use?

io.github.jeet-dhandha/agent-webbridge scores 78 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 io.github.jeet-dhandha/agent-webbridge MCP server expose?

io.github.jeet-dhandha/agent-webbridge exposes 16 tools: browser_status, browser_navigate, browser_snapshot, browser_click, browser_trusted_click, and 11 more. Their descriptions and schemas cost roughly 2,533 tokens of context every time the server is loaded.

### Is the io.github.jeet-dhandha/agent-webbridge MCP server still maintained?

io.github.jeet-dhandha/agent-webbridge 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 io.github.jeet-dhandha/agent-webbridge MCP server under?

io.github.jeet-dhandha/agent-webbridge 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/agent-webbridge
- Socket report: https://socket.dev/npm/package/agent-webbridge
- Repository: https://github.com/jeet-dhandha/agent-webbridge
- Changelog RSS feed: https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge.xml
- Changelog JSON feed: https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge.json
- HTML version of this page: https://verifymcp.io/servers/jeet-dhandha-agent-webbridge/agent-webbridge