← 开源
kagisearch

kagimcp

The Official Model Context Protocol (MCP) server for Kagi Search & other tools.

AI EngineeringGive agent toolsPython
在 GitHub 打开
增长势头
+224 小时新增 Star+0.4%
544
Star
49
Fork
+9
本周
7
贡献者
创建于 2024-12-13 · 更新于 2026-10-16 · 今日第 4280 名
主要开发者

kagisearch/kagimcp 排名第 4,280,有 544 个星标。

同类别比较

kagisearch/kagimcp 排名第 4,280,此处有 8 个同类条目;jalaalrd/anti-ai-slop-writing 排名第 4,279。

同类别比较
名称排名指标
ZJU-LLMs/Agent-Kernel4,274
语言:
Python
星标数:
550
分叉数:
52
简介:
A MicroKernel Multi-Agents System Framework for Adaptive Social Simulation Powered by LLMs
czm15053/write-notes-like-deepseek4,275
语言:
HTML
星标数:
550
分叉数:
46
avansaber/seo-monster4,276
语言:
Python
星标数:
548
分叉数:
83
简介:
SEOMonster: a local, user-credential-driven MCP server for SEO workflows over Google Search Console, GA4, PageSpeed Insights, and Cloudflare. Read-first, open source.
jalaalrd/anti-ai-slop-writing4,279
星标数:
544
分叉数:
53
简介:
AI writing skill that eliminates detectable AI patterns. Works with Claude Code, Codex, Cursor, Gemini CLI, and 8+ other agents.
WhiteNightShadow/camoufox-reverse-mcp4,282
语言:
Python
星标数:
540
分叉数:
104
简介:
🔧 反检测浏览器 MCP Server | 35 个 JS 逆向工具 + C++ 引擎层 JSVMP 指纹追踪
DavidCarliez/trustmebro4,290
语言:
Go
星标数:
532
分叉数:
48
简介:
Bypass llm guardrails by confusing it with fabricated tool output.
yanliudesign/offer-toolkit-skill4,293
语言:
HTML
星标数:
531
分叉数:
59
motherduckdb/mcp-server-motherduck4,295
语言:
Python
星标数:
529
分叉数:
87
简介:
Local MCP server for DuckDB and MotherDuck
关键参数与规格
关键参数与规格
参数值
语言Python
类别AI Engineering
子类别Give agent tools
关键词Agent-ready APIs, Data extraction, MCP, Retrieval
星标数544
分叉数49
贡献者7
创建日期2024-12-13
更新日期2026-10-16
1 天星标数2
7 天星标数9
排名4,280
上次排名17,167
排名变化12,887
7 天星标排名4,319
上次 7 天星标排名4,914
7 天星标排名变化595
总星标排名18,124
上次总星标排名18,161
总星标排名变化37
国家USA
README 摘录

Kagi MCP Server An MCP server backed by the Kagi API. It exposes search and extraction tools to MCP-compatible clients. Tools ** ** - web, news, videos, podcasts, and image search with optional page extracts, filters, and Kagi lenses. ** ** - fetch a page's full content as markdown. **Note:** The previous and tools have been removed. Both are planned to return in a future release. Point any HTTP-capable MCP client at it and authenticate with your Kagi API key. OAuth2 isn't supported yet

(it's on our roadmap), so for now grab your API key from the dashboard and pass it via HTTP authentication. Example with Claude Code: Prefer to run it yourself? See Client Setup for the local install, or Self-Hosting to host the HTTP server on your own infrastructure. Install : Windows: Client Setup Codex CLI Codex writes MCP configuration to . MacOS/Linux: Windows: Then in your Claude Desktop config (found through Settings -> Developer -> Edit Config): Claude Code Smithery Kiro Add to

your Kiro MCP config file ( for global, or for project-scoped) using the same JSON as Claude Desktop. See the Kiro MCP documentation for more details. | Max retry attempts after the first request. | Comma-separated search params to hide from the LLM-facing schema. Hideable search params: Example: Local Development Run locally over stdio: Run with streamable HTTP transport: Self-Hosting HTTP mode is multi-tenant: each request supplies its API key via the header instead of a server-wide

env var, so one instance can serve multiple users. The repo ships a that installs a pinned from PyPI and runs it in HTTP mode. The container respects so it works on any platform that injects one (Railway, Render, Cloud Run, Fly. Build and run locally: Smoke test: To bump the version in production, edit the pin in the and redeploy. Debugging Inspect the published package: Inspect a local checkout: The inspector is usually available at . Prerelease Instructions If using

a prerelease build, the same installation instructions apply, but use instead of (replace with whatever version you're wanting to install).

常见问题

记录了多少星标和分叉?

kagisearch/kagimcp 有 544 个星标和 49 个分叉。

仓库排名是多少?

kagisearch/kagimcp 排名第 4,280。

记录的语言和类别是什么?

kagisearch/kagimcp 有 544 个星标,语言为 Python,类别为 AI Engineering。

哪个仓库排名最接近?

kagisearch/kagimcp 与 jalaalrd/anti-ai-slop-writing 比较,后者排名第 4,279,有 544 个星标。

README

Kagi MCP Server

An MCP server backed by the Kagi API. It exposes search and extraction tools to MCP-compatible clients.

Tools

  • kagi_search_fetch - web, news, videos, podcasts, and image search with optional page extracts, filters, and Kagi lenses.
  • kagi_extract - fetch a page's full content as markdown.

Note: The previous kagi_fastgpt and kagi_summarizer tools have been removed. Both are planned to return in a future release.

Hosted Server

We run a hosted MCP server at https://mcp.kagi.com/mcp — no install required. Point any HTTP-capable MCP client at it and authenticate with your Kagi API key.

OAuth2 isn't supported yet (it's on our roadmap), so for now grab your API key from the dashboard and pass it via Bearer HTTP authentication.

Example with Claude Code:

claude mcp add kagi https://mcp.kagi.com/mcp --transport http --header "Authorization: Bearer $(read -sp 'API key: ' k; echo $k)" --scope user

Prefer to run it yourself? See Client Setup for the local uvx install, or Self-Hosting to host the HTTP server on your own infrastructure.

Requirements

  • A Kagi API key in KAGI_API_KEY.
  • uv for the recommended uvx install path.

Install uv:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Client Setup

Codex CLI

codex mcp add kagi --env KAGI_API_KEY= -- uvx kagimcp

Codex writes MCP configuration to ~/.codex/config.toml.

Claude Desktop

Install uv first.

MacOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Then in your Claude Desktop config (found through Settings -> Developer -> Edit Config):

{
  "mcpServers": {
    "kagi": {
      "command": "uvx",
      "args": ["kagimcp"],
      "env": {
        "KAGI_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

Claude Code

claude mcp add kagi -e KAGI_API_KEY="YOUR_API_KEY_HERE" -- uvx kagimcp

Smithery

npx -y @smithery/cli install kagimcp --client claude

Kiro

Add to your Kiro MCP config file (~/.kiro/settings/mcp.json for global, or .kiro/settings/mcp.json for project-scoped) using the same mcpServers JSON as Claude Desktop. See the Kiro MCP documentation for more details.

OpenCode

Edit the OpenCode configuration file in ~/.config/opencode/opencode.json and add the following:

{
  "mcp": {
    "kagi": {
      "type": "local",
      "command": ["uvx", "kagimcp"],
      "enabled": true,
      "environment": {
        "KAGI_API_KEY": ""
      }
    }
  }
}

Usage Examples

  • Search: Who was Time's 2024 person of the year?
  • Extract: extract the full content of https://en.wikipedia.org/wiki/Model_Context_Protocol

Configuration

Environment variable Description
KAGI_API_KEY Required Kagi API key.
FASTMCP_LOG_LEVEL Logging level, for example ERROR.
KAGI_SEARCH_TIMEOUT Search timeout in seconds. Defaults to 10.
KAGI_EXTRACT_TIMEOUT Extract timeout in seconds. Defaults to 30.
KAGI_MAX_RETRIES Max retry attempts after the first request. Defaults to 2; set 0 to disable retries.
KAGI_HIDDEN_PARAMS Comma-separated search params to hide from the LLM-facing schema.

Hideable search params:

workflow, extract_count, limit, include_domains, exclude_domains, time_relative, after, before, file_type, lens_id

Example:

KAGI_HIDDEN_PARAMS="extract_count,after,before,time_relative,include_domains,exclude_domains"

Local Development

git clone https://github.com/kagisearch/kagimcp.git
cd kagimcp
uv sync

Run locally over stdio:

KAGI_API_KEY= uv run kagimcp

Run with streamable HTTP transport:

KAGI_API_KEY= uv run kagimcp --http --host 0.0.0.0 --port 8000

Self-Hosting

HTTP mode is multi-tenant: each request supplies its API key via the Authorization: Bearer header instead of a server-wide env var, so one instance can serve multiple users. The repo ships a Dockerfile that installs a pinned kagimcp from PyPI and runs it in HTTP mode. The container respects $PORT so it works on any platform that injects one (Railway, Render, Cloud Run, Fly.io, etc.).

Build and run locally:

docker build -t kagimcp-hosted .
docker run --rm -p 8000:8000 kagimcp-hosted

Smoke test:

curl -sL http://127.0.0.1:8000/mcp -X POST \
  -H "authorization: Bearer $KAGI_API_KEY" \
  -H "content-type: application/json" \
  -H "accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

To bump the version in production, edit the pin in the Dockerfile and redeploy.

Debugging

Inspect the published package:

npx @modelcontextprotocol/inspector uvx kagimcp

Inspect a local checkout:

npx @modelcontextprotocol/inspector uv --directory /ABSOLUTE/PATH/TO/kagimcp run kagimcp

The inspector is usually available at http://localhost:5173.

Prerelease Instructions

If using a prerelease build, the same installation instructions apply, but use uvx --prerelease allow --from kagimcp==1.0.0rc2 kagimcp instead of uvx kagimcp (replace 1.0.0rc2 with whatever version you're wanting to install).

排名相邻
同类项目