Skip to main content
Glama
deenesjakoruzh
chenxuyaun

free-web-mcp

by chenxuyaun
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

free_web_mcp

Free Web MCP — 무료 MCP Server로, Claude / Cursor / 기타 MCP Client에 웹 검색(DuckDuckGo, API Key 불필요)과 웹 페이지 크롤링 + 본문 추출 도구를 제공합니다.

자세한 요구 사항은 docs/free_web_mcp.md를 참조하세요.

도구

도구

설명

web_search(query, max_results=5)

DuckDuckGo 검색, {title,url,snippet,source} 목록 반환

web_fetch(url, rendered=False)

웹 페이지를 크롤링하고 본문을 추출하여 {url,title,content,text_length} 반환; rendered=True이면 headless Chromium 구동(RENDER_ENABLED=true 필요)

web_search_and_fetch(query, max_results=5, rendered=False)

검색 후 Top N 결과의 본문을 하나씩 크롤링

Related MCP server: qsearch

빠른 시작

uv sync

stdio 모드(로컬 클라이언트)

uv run free-web-mcp                # 或: uv run python -m free_web_mcp

Cursor / Claude Desktop 구성 예시:

{
  "mcpServers": {
    "free-web-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/free_web_mcp", "free-web-mcp"]
    }
  }
}

HTTP 모드(원격 배포)

uv run free-web-mcp --transport http   # 监听 .env 中的 HOST/PORT，默认 0.0.0.0:8000
curl http://localhost:8000/health      # -> {"status":"ok","service":"free-web-mcp"}

MCP streamable-http 엔드포인트: http://<host>:8000/mcp.

Docker

# 默认：仅 HTTP 抓取路径
docker build -f docker/Dockerfile -t free-web-mcp .
docker run --rm -p 8000:8000 free-web-mcp

# 启用 headless 渲染（构建时下载 Chromium ~150MB；首次启动约 1-3s）
docker build -f docker/Dockerfile --build-arg INSTALL_PLAYWRIGHT_BROWSERS=true -t free-web-mcp:render .
docker run --rm -p 8000:8000 -e RENDER_ENABLED=true free-web-mcp:render

구성(.env)

.env.example을 .env로 복사한 후 필요에 따라 수정하세요. .env는 저장소에 포함되지 않습니다.

변수

기본값

설명

HOST / PORT

0.0.0.0 / 8000

HTTP 수신 주소

LOG_LEVEL

INFO

로그 레벨

HTTP_TIMEOUT

30

크롤링 타임아웃(초)

MAX_CONTENT_LENGTH

5000000

응답 본문 크기 상한(바이트)

SEARCH_MAX_RESULTS

10

max_results의 상한

RENDER_ENABLED

false

headless 브라우저(Playwright + Chromium) 활성화 여부

RENDER_TIMEOUT

30

렌더링 타임아웃(초)

RENDER_MAX_BYTES

5000000

렌더링 후 응답 본문 크기 상한

개발

uv run pytest                  # 测试
uv run ruff check .            # lint
uv run mypy                    # 类型检查（strict）

아키텍처

엄격한 계층 구조: MCP Tool 계층 → Service 계층(SearchService/FetchService) → Provider/HTTP 계층. 검색 소스는 플러그인 방식으로 교체 가능(현재 구현: DuckDuckGoProvider), HTTP는 WebClient로 통일, 본문 추출은 trafilatura 우선, BeautifulSoup 폴백. 자세한 내용은 AGENTS.md를 참조하세요.

Available Tools

3 tools
web_fetchB

Fetch a webpage and extract its main readable content.

Set rendered=True to drive a headless browser (requires RENDER_ENABLED=true).

ParametersJSON Schema
NameRequiredDescriptionDefault
urlYes
renderedNo

Output Schema

ParametersJSON Schema
NameRequiredDescription

No output parameters

TDQS

B3.4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full transparency burden. It usefully discloses that rendered=True drives a headless browser and requires RENDER_ENABLED=true, going beyond the schema. However, it does not explain the default non-rendered behavior, failure modes, or other behavioral constraints.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two short sentences with the main purpose front-loaded and the optional rendering mode stated compactly. Every sentence earns its place and there is no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple two-parameter fetch tool with an output schema, the description covers the core behavior and the one non-obvious mode. It is slightly incomplete in that it omits explicit sibling routing and default-mode behavior, but it is sufficient for basic invocation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate. It adds real meaning for rendered by explaining the headless-browser behavior and the RENDER_ENABLED requirement, but the url parameter is only implicitly covered by 'Fetch a webpage' and gets no format or constraint detail.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb ('Fetch') and resource ('a webpage') and adds the key value extra ('extract its main readable content'), making the core purpose clear. It does not explicitly differentiate from the sibling web_search_and_fetch, which may also involve fetching pages.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives no guidance about when to use this tool instead of web_search or web_search_and_fetch. The rendered=True note is a configuration instruction, not a usage-selection guideline, and no alternatives or exclusions are mentioned.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

web_search_and_fetchA

Search the web, then fetch and extract text from each result URL.

Set rendered=True to drive a headless browser (requires RENDER_ENABLED=true).

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYes
renderedNo
max_resultsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription

No output parameters

TDQS

A3.6/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full behavioral disclosure burden. It does disclose that rendered=True drives a headless browser and requires RENDER_ENABLED=true, and that it fetches from each result URL. However, it omits potential issues like rate limits, latency, fetch failures, or how many URLs are actually processed.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two concise sentences with the core operation front-loaded and the conditional rendered caveat placed after. Every sentence earns its place and there is no redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The output schema covers return shape, so that omission is acceptable. However, for a compound tool with no annotations, the description leaves meaningful gaps around max_results behavior, fetch error handling, and operational implications of fetching multiple URLs.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, and the description only explains the rendered parameter. The query and max_results parameters receive no semantic explanation beyond their names, and max_results in particular is non-obvious regarding how it limits the fetch stage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific compound action: search the web, then fetch and extract text from each result URL. This clearly distinguishes it from the sibling tools web_search and web_fetch, which each cover only one stage of the workflow.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies this tool is for combined search-and-fetch workflows, but it never explicitly names the sibling tools or states when to prefer a single-stage tool instead. The usage guidance is left mostly to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 3 tool updatesv0.1.0
    • First observedweb_fetch
    • First observedweb_search
    • First observedweb_search_and_fetch
Deploy Server
Let your AI agent see the webWeb-to-screenshot MCP for Claude, Cursor and Codex. No cookie banners. Free to start.snap-render.comAd

TDQS

A3.7/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a clearly distinct purpose: web_search returns results, web_fetch retrieves a single page, and web_search_and_fetch explicitly combines both. There is no meaningful overlap between the three.

Naming Consistency5/5

All tool names follow the consistent web_<action> pattern using lowercase snake_case. The names clearly indicate their operation, including the compound web_search_and_fetch.

Tool Count5/5

Three tools is a reasonable, well-scoped size for a simple web retrieval server. Each tool earns its place, and the combined search-and-fetch adds convenience without bloating the surface.

Completeness4/5

The core search-and-fetch workflow is fully covered, with an optional rendered mode for JavaScript-heavy pages. A minor gap is lack of a tool for raw HTML or structured page extraction, but most basic browsing needs are addressed.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • qsearch

    theYahia
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables AI agents to perform web searches with full content retrieval and multi-engine provenance, including trust scoring and local corpus persistence, via MCP integration.
    1 npm
    3
    Apache 2.0

  • jiro

    DevAnimecx
    A
    license
    Not graded
    quality
    A
    maintenance
    Enables AI agents to perform live web searches across 9 engines, scrape web pages into clean formats, and run agentic research with citations via MCP.
    98 PyPI
    3
    MIT

  • CrawlEyes

    waiky-github
    A
    license
    B
    quality
    B
    maintenance
    Enables AI agents to perform web searches and extract full-text content from web pages via standard MCP tools, with fallback search and semantic reranking.
    2
    2
    MIT