Devin Search MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DEVIN_PATH | No | Optional path to the Devin Desktop executable if installed in a non-standard location, e.g. E:\Program\devin\resources\app\extensions\windsurf\devin\bin\devin.exe. Not required if Devin Desktop is installed in the default location and already logged in. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| devin_web_searchA | 极速全网实时检索 (直连 Devin 官方网关)。用自然语言查最新技术文档、库版本特性、报错与开源方案，亚秒级返回结构化标题、网址与摘要。 |
| devin_web_fetchB | 轻量抓取指定网页的正文内容，自动剔除导航、广告与脚本，输出整洁的 Markdown 或纯文本。 |
| extract_devin_keyB | 自动提取本地 Devin / Windsurf 登录凭证 (对齐 fast-context-mcp 的 extract_windsurf_key)。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
Each tool targets a distinct action: web search, web content fetching, and credential extraction. Boundaries are clear, and the credential tool does not overlap with the web retrieval tools.
Two tools use a devin_web_ prefix and noun-based structure, while extract_devin_key uses a different verb-first pattern without the web prefix. All names are snake_case and readable, but the convention is not uniform.
Three tools is a compact set appropriate for a focused search server, though the credential extraction tool is peripheral to the core search/fetch purpose. No bloat, and each tool has a clear role.
Search and fetch cover the primary web retrieval lifecycle, and key extraction supports authenticated access. Minor gaps like batch fetch, date filters, or search operators exist, but the core operations are present.