# Personae (npm · @leek-emperor/personae-mcp)

Control specific, already-open account-isolated Personae browser identities with MCP.

- Trust score: 83/100 (high trust)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-20

## Components

- npm · `@leek-emperor/personae-mcp`: 83/100 (this document), [markdown](https://verifymcp.io/servers/leek-emperor-personae/leek-emperor-personae-mcp.md), [page](https://verifymcp.io/servers/leek-emperor-personae/leek-emperor-personae-mcp)

## Channel facts

- Registry: `npm`
- Package: `@leek-emperor/personae-mcp`
- Version: `0.2.1`
- 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 2026-09-20.

- **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.
  - No production dependencies, so there is no dependency health to assess.
- **Provenance & Transparency**: 97/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to leek-emperor/Personae).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 11 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 80/100
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 1266 tokens (~97/item across 13 items; 13 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 40/100
  - Stability observed for 12 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 88/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 65% 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; "eval_js" implies "eval" 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**: 60/100
  - Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28.

## Install

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

Personae runs locally as an npm package, launched with npx -y @leek-emperor/personae-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 leek-emperor-personae -- npx -y @leek-emperor/personae-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "leek-emperor-personae": {
      "command": "npx",
      "args": [
        "-y",
        "@leek-emperor/personae-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "leek-emperor-personae": {
      "command": "npx",
      "args": [
        "-y",
        "@leek-emperor/personae-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add leek-emperor-personae -- npx -y @leek-emperor/personae-mcp
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add leek-emperor-personae --command npx --arg -y --arg @leek-emperor/personae-mcp
```

### Hermes

```yaml
mcp_servers:
  leek-emperor-personae:
    command: "npx"
    args: ["-y", "@leek-emperor/personae-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "leek-emperor-personae": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "@leek-emperor/personae-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add leek-emperor-personae -t stdio -c npx -a -y @leek-emperor/personae-mcp
```

### Other

```json
{
  "mcpServers": {
    "leek-emperor-personae": {
      "command": "npx",
      "args": [
        "-y",
        "@leek-emperor/personae-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.

### 2026-09-19 (score 83, +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.

### 2026-09-16 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-15 (score 81, +3)

- [functional improvement] Stability: unverified → 0.23

### 2026-09-09 (score 78, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-09-08 (score 63)

First indexed and scored.

## MCP tools (13)

### `load_skill` (~188 tokens)

Load agent-browser 官方命令语法说明，内容随本机捆绑的二进制版本，保证 flag 真实存在。**在用 act 工具手写 agent-browser 命令前必须先调用本工具** —— 它的语法在版本间会变（本机版本就没有网上文档里的 `tab --url` 和 `--pin-tab`），凭记忆猜必然失败。注意：本 app 的多身份窗口应优先用 snapshot/click/fill/navigate 等专用工具操作，它们已处理好「身份→窗口」定位；只有需要专用工具没覆盖的能力时才手写命令。

Input parameters:

- `section` (string): 可选：只关心某方面时填关键词过滤，如 snapshot / click / tab / eval / screenshot / batch。留空返回全部语法。

### `list_identities` (~59 tokens)

列出所有浏览器身份（partition）及其 CDP targetId、是否已打开、当前页面。任何操作前先调用此工具确认身份状态 —— 这是把 partition 对应到 CDP target 的唯一途径。

### `open_identity` (~50 tokens)

打开（或聚焦）指定身份的浏览器窗口。窗口打开后才会生成 targetId，agent 才能操作它。

Input parameters:

- `identity` (string, required): 身份 id 或名称

### `snapshot` (~189 tokens)

读取指定身份窗口的可交互元素快照（accessibility tree + @eN refs）。会自动定位到该身份的窗口，不会读到别的身份。注意：ref 只在单次调用内有效，click/fill 等工具会各自重新 snapshot，因此你只需把「元素的可读名称」记住，让后续工具自己按名称定位；若要按 ref 操作，请用 act 工具在一次调用内完成。若要读该身份下弹出的子窗口（OAuth/分享等），传 target=子窗口的 targetId（见 list_identities）。

Input parameters:

- `compact` (boolean): 精简输出，默认 true
- `identity` (string, required): 身份 id 或名称
- `target` (string): 可选：该身份下子窗口的 targetId 或序号，不传则操作主窗口

### `navigate` (~74 tokens)

在指定身份的窗口内导航到 URL。跳转被限制在该窗口内，不会新开窗口或外跳系统浏览器。

Input parameters:

- `identity` (string, required)
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口
- `url` (string, required)

### `click` (~120 tokens)

点击指定身份窗口中的元素。按元素可读名称定位（推荐），或直接给 @eN ref —— 本工具内部会先重新 snapshot 再点击，所以 ref 用的是最新一次快照的编号。

Input parameters:

- `identity` (string, required)
- `ref` (string): @eN 形式的元素 ref
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口
- `text` (string): 元素可读文本，如「百度一下」；与 ref 二选一

### `fill` (~117 tokens)

清空并填入文本。按元素可读名称/label 定位（推荐），或给 @eN ref（内部会先重新 snapshot）。

Input parameters:

- `identity` (string, required)
- `label` (string): 输入框的 label 或占位文本；与 ref 二选一
- `ref` (string): @eN 形式的元素 ref
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口
- `text` (string, required): 要填入的内容

### `press` (~65 tokens)

发送按键，如 Enter、Tab、Control+a。作用于该身份窗口当前焦点。

Input parameters:

- `identity` (string, required)
- `key` (string, required)
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

### `act` (~143 tokens)

在一次调用内对同一身份连续执行多个 agent_browser 命令，ref 在这些命令之间有效。适合「snapshot → 点这个 → 填那个 → 回车」这种多步流程。命令用 agent_browser 语法，如 ["snapshot -i", "fill @e35 关键词", "press Enter"]。会自动前置 tab 切换。

Input parameters:

- `commands` (array, required): agent_browser 命令字符串数组，按顺序在同一进程内执行
- `identity` (string, required)
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

### `get_text` (~78 tokens)

读取元素文本。可传 @eN ref（内部会先重新 snapshot）或 CSS 选择器。

Input parameters:

- `identity` (string, required)
- `ref` (string, required): @eN ref 或 CSS 选择器
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

### `get_url` (~47 tokens)

读取该身份窗口当前的 URL。

Input parameters:

- `identity` (string, required)
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

### `screenshot` (~68 tokens)

截图指定身份的窗口，保存到给定绝对路径。

Input parameters:

- `identity` (string, required)
- `path` (string, required): 输出文件绝对路径，.png
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

### `eval_js` (~68 tokens)

在指定身份窗口内执行 JavaScript 并返回结果。用于读取页面状态、验证隔离等。

Input parameters:

- `expression` (string, required)
- `identity` (string, required)
- `target` (string): 可选：子窗口 targetId 或序号，不传则操作主窗口

## Diagnostics

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

## Score history

- 2026-09-20: 83
- 2026-09-19: 83
- 2026-09-18: 82
- 2026-09-17: 82
- 2026-09-16: 82
- 2026-09-15: 81
- 2026-09-14: 78
- 2026-09-13: 78
- 2026-09-12: 78
- 2026-09-11: 78
- 2026-09-10: 78
- 2026-09-09: 78
- 2026-09-08: 63

## Common questions

### What is the Personae MCP server?

Personae is an MCP server listed in the public MCP registry as io.github.leek-emperor/personae. Control specific, already-open account-isolated Personae browser identities with MCP. This page covers its npm package (@leek-emperor/personae-mcp).

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

Personae scores 83 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. It declares no install or post-install scripts. Its build provenance is signed and verified. 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 Personae MCP server expose?

Personae exposes 13 tools: load_skill, list_identities, open_identity, snapshot, navigate, and 8 more. Their descriptions and schemas cost roughly 1,266 tokens of context every time the server is loaded.

### Is the Personae MCP server still maintained?

Personae is still listed as active in the MCP registry. We last reached this channel on 20 September 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 Personae MCP server under?

Personae 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/@leek-emperor/personae-mcp
- Socket report: https://socket.dev/npm/package/@leek-emperor/personae-mcp
- Repository: https://github.com/leek-emperor/Personae
- Changelog RSS feed: https://verifymcp.io/servers/leek-emperor-personae/leek-emperor-personae-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/leek-emperor-personae/leek-emperor-personae-mcp.json
- HTML version of this page: https://verifymcp.io/servers/leek-emperor-personae/leek-emperor-personae-mcp