# DeepSeek Search Everywhere (pypi · deepseek-search-everywhere-mcp)

Web search, page fetching and multi-source research via the DeepSeek Search & Fetch API.

- Trust score: 75/100 (medium)
- Change this week: −1
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- pypi · `deepseek-search-everywhere-mcp`: 75/100 (this document), [markdown](https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp.md), [page](https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp)

## Channel facts

- Registry: `pypi`
- Package: `deepseek-search-everywhere-mcp`
- Version: `2.0.2`
- 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.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - 0 of 28 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to radoxtech/deepseek-search-everywhere-mcp).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 3 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 58/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1885 tokens (~269/item across 7 items; 7 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: the tool surface changed between 1.0.0 and 2.0.2: 3 tool removals, 0 breaking changes, 4 additions.
- **Tool Coverage**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **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; "deepseek_deep_research" implies "execute" and declares readOnlyHint instead, contradicting what its own name says it does.
  - An AI judge read all 7 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 DeepSeek Search Everywhere MCP server?

DeepSeek Search Everywhere runs locally as a PyPI package, launched with uvx deepseek-search-everywhere-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 radoxtech-deepseek-search-everywhere-mcp -- uvx deepseek-search-everywhere-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "radoxtech-deepseek-search-everywhere-mcp": {
      "command": "uvx",
      "args": [
        "deepseek-search-everywhere-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "radoxtech-deepseek-search-everywhere-mcp": {
      "command": "uvx",
      "args": [
        "deepseek-search-everywhere-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add radoxtech-deepseek-search-everywhere-mcp -- uvx deepseek-search-everywhere-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "radoxtech-deepseek-search-everywhere-mcp": {
      "type": "local",
      "command": [
        "uvx",
        "deepseek-search-everywhere-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add radoxtech-deepseek-search-everywhere-mcp --command uvx --arg deepseek-search-everywhere-mcp
```

### Hermes

```yaml
mcp_servers:
  radoxtech-deepseek-search-everywhere-mcp:
    command: "uvx"
    args: ["deepseek-search-everywhere-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "radoxtech-deepseek-search-everywhere-mcp": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "deepseek-search-everywhere-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add radoxtech-deepseek-search-everywhere-mcp -t stdio -c uvx -a deepseek-search-everywhere-mcp
```

### Other

```json
{
  "mcpServers": {
    "radoxtech-deepseek-search-everywhere-mcp": {
      "command": "uvx",
      "args": [
        "deepseek-search-everywhere-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.

### 2026-11-01 (score 75, −2)

- [security regression] Tool safety: pass → unverified
- [security regression] Stability: 0.10 → unverified
- [security regression] Stability: 0.10 → fail
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] Schema quality: 558 → 269
- [functional improvement] Schema quality: 558 → 230
- [functional improvement] Schema quality: 558 → 246
- [functional improvement] Schema quality: 558 → 287
- [functional improvement] Schema quality: 558 → 289
- [functional] First check of Schema quality: unverified
- [functional] Schema quality: excellent → good
- [functional] Destructive annotations: pass → 0
- [functional] Package version: 1.0.0 → 2.0.2
- [functional] Package version: 1.0.0 → 2.0.1
- [functional] Package version: 1.0.0 → 1.3.1
- [functional] Package version: 1.0.0 → 1.2.0
- [functional] Package version: 1.0.0 → 1.1.0

### 2026-10-30 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 76)

First indexed and scored.

## MCP tools (7)

### `deepseek_search` (~448 tokens)

Search the web with DeepSeek index and return page text, snippets, and metadata.

Primary search and retrieval tool for AI agents. Queries DeepSeek's search index directly
to return ranked results with snippets (need_content=False) or indexed full page text
(default need_content=True). Use for standard search, targeted queries, and discovery.
Unlike deepseek_research (auto-fetches stub pages), this tool only returns indexed content.

Args:
    query: Search query, max 128 chars (supports site:, -site:, "phrase", after:, before:).
    count: Results to return, 1-50 (default 5).
    ori_question: Natural question, max 128 chars; improves ranking.
    need_content: Full text inline (default True; False for snippets).
    lang: Language filter (e.g. 'es', 'de', 'ja').
    max_chars_per_result: Cap per result (default 4000).
    max_total_chars: Budget across results (default 30000).
    dedupe: Collapse duplicate results (default True).
    rewrite: Rewrite query via deepseek-flash (default False).
    after: Date on or after (YYYY-MM-DD/YYYY).
    before: Date on or before (YYYY-MM-DD/YYYY).
    include_domains: Restrict domains (e.g. ['arxiv.org']).
    exclude_domains: Exclude domains (e.g. ['wikipedia.org']).
    refresh: Bypass response cache (default False).
    summary_chars: Summary cap per result without content.
    strip_boilerplate: Strip site UI text (default True).

Input parameters:

- `after`
- `before`
- `count` (integer)
- `dedupe` (boolean)
- `exclude_domains`
- `include_domains`
- `lang`
- `max_chars_per_result` (integer)
- `max_total_chars` (integer)
- `need_content` (boolean)
- `ori_question`
- `query` (string, required)
- `refresh` (boolean)
- `rewrite` (boolean)
- `strip_boilerplate` (boolean)
- `summary_chars`

### `deepseek_fetch` (~254 tokens)

Fetch one known URL and return its title and Markdown content.

Retrieves full Markdown content for a known URL. Use when target URL is known.
For multiple URLs use deepseek_fetch_many; for structured data use deepseek_extract;
for search use deepseek_search. Returns 404 for PDFs/Reddit (use deepseek_search instead).

Args:
    url: Absolute http(s) URL, max 8192 characters.
    clean: Remove Markdown backslash escapes (default True).
    max_chars: Content cap after cleaning (default 20000).
    query: Query to extract ranked relevant snippets.
    section: 1-based index to read a specific document section.
    max_snippets: Max ranked snippets when query is set (default 3, 1-5).
    refresh: Bypass cache and fetch page again (default False).
    strip_boilerplate: Remove site UI lines from text (default True).

Input parameters:

- `clean` (boolean)
- `max_chars` (integer)
- `max_snippets` (integer)
- `query`
- `refresh` (boolean)
- `section`
- `strip_boilerplate` (boolean)
- `url` (string, required)

### `deepseek_fetch_many` (~264 tokens)

Fetch up to 10 known URLs concurrently and return each page's title and Markdown content.

Fetch multiple known URLs in parallel. Use when target URLs are known. Do not use
for a single URL (use deepseek_fetch), nor for search/discovery (use deepseek_search).
Skips blanks, deduplicates, and preserves order.

Args:
    urls: 1-10 absolute http(s) URLs, max 8192 characters each.
    max_chars_per_url: Content cap per page (default 4000).
    max_total_chars: Total content budget across all pages (default 20000).
    clean: Remove Markdown backslash escapes (default True).
    query: Query to extract ranked snippets across fetched pages.
    refresh: Bypass cache and fetch pages again (default False).
    strip_boilerplate: Remove site UI lines from text (default True).
    deadline: Max seconds before cancelling pending fetches (default 120s).

Input parameters:

- `clean` (boolean)
- `deadline`
- `max_chars_per_url` (integer)
- `max_total_chars` (integer)
- `query`
- `refresh` (boolean)
- `strip_boilerplate` (boolean)
- `urls` (array, required)

### `deepseek_research` (~372 tokens)

Single-pass source gathering: search with page content and auto-fetch stub pages.

Gathers candidate sources for research. It does not iterate, refine the query or summarise:
there is exactly one search and sources return raw content for the caller to synthesise.

Unlike deepseek_search (index-only), deepseek_research auto-fetches stub results in parallel
for deep source gathering. Use when research requires complete documents rather than snippets.

Args:
    question: Search query, max 128 characters.
    max_pages: Results to use, 1-50 (default 5).
    per_page_chars: Cap per source (default 6000).
    total_chars: Budget across sources (default 30000).
    after: Date on or after (YYYY-MM-DD/YYYY).
    before: Date on or before (YYYY-MM-DD/YYYY).
    include_domains: Restrict domains (e.g. ['arxiv.org']).
    exclude_domains: Exclude domains (e.g. ['wikipedia.org']).
    lang: Language filter (e.g. 'pl', 'es', 'ja').
    rewrite: Translate query via deepseek-flash (default False).
    dedupe: Collapse duplicate results (default True).
    refresh: Bypass response cache (default False).
    strip_boilerplate: Strip site UI text (default True).

Input parameters:

- `after`
- `before`
- `dedupe` (boolean)
- `exclude_domains`
- `include_domains`
- `lang`
- `max_pages` (integer)
- `per_page_chars` (integer)
- `question` (string, required)
- `refresh` (boolean)
- `rewrite` (boolean)
- `strip_boilerplate` (boolean)
- `total_chars` (integer)

### `deepseek_deep_research` (~242 tokens)

Execute multi-angle research on a topic and generate a cited briefing report.

Decomposes topic into subqueries, retrieves sources concurrently, and synthesizes
a cited report [1], [2]. Use for deep topic synthesis. Do not use for simple lookups
(use deepseek_search/deepseek_research) or reading known URLs (use deepseek_fetch).

Args:
    topic: Research subject, max 500 chars.
    max_queries: Subqueries to run (2-4, default 3).
    max_pages_per_query: Sources per subquery (1-5, default 3).
    lang: Language filter (e.g. 'en', 'pl').
    dedupe: Collapse duplicates across subqueries (default True).
    refresh: Bypass response cache (default False).
    strip_boilerplate: Strip site UI text (default True).

Input parameters:

- `dedupe` (boolean)
- `lang`
- `max_pages_per_query` (integer)
- `max_queries` (integer)
- `refresh` (boolean)
- `strip_boilerplate` (boolean)
- `topic` (string, required)

### `deepseek_extract` (~233 tokens)

Extract structured JSON from a web page conforming to target fields or a schema.

Extracts specific entities or key-value data without returning full page text.
Returns extracted JSON under 'data'. Do not use for general reading (use deepseek_fetch),
nor for search or discovery (use deepseek_search).

Args:
    url: Absolute web page URL to extract from.
    fields: Target attribute names (e.g. ['price', 'specs']), exclusive with schema.
    schema: JSON Schema defining target structure, exclusive with fields.
    instruction: Optional extraction guidance.
    clean: Remove Markdown backslash escapes (default True).
    max_chars: Content cap before extraction (default 40000).
    refresh: Bypass cache and fetch page again (default False).
    strip_boilerplate: Remove site UI lines from text (default True).

Input parameters:

- `clean` (boolean)
- `fields`
- `instruction`
- `max_chars` (integer)
- `refresh` (boolean)
- `schema`
- `strip_boilerplate` (boolean)
- `url` (string, required)

### `deepseek_check_balance` (~72 tokens)

Check DeepSeek account balance (CNY/USD) and in-memory session telemetry.

Verify account funds and monitor session usage (requests, cache hits, estimated cost).
Do not use for search or retrieval (use deepseek_search or deepseek_fetch). Requires
DEEPSEEK_API_KEY. Takes no arguments.

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp#diagnostics

## Score history

- 2026-11-05: 75
- 2026-11-04: 75
- 2026-11-03: 75
- 2026-11-02: 75
- 2026-11-01: 75
- 2026-10-31: 77
- 2026-10-30: 77
- 2026-10-29: 76
- 2026-10-28: 76

## Common questions

### What is the DeepSeek Search Everywhere MCP server?

DeepSeek Search Everywhere is an MCP server listed in the public MCP registry as io.github.radoxtech/deepseek-search-everywhere-mcp. Web search, page fetching and multi-source research via the DeepSeek Search & Fetch API. This page covers its PyPI package (deepseek-search-everywhere-mcp).

### Is the DeepSeek Search Everywhere MCP server safe to use?

DeepSeek Search Everywhere scores 75 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. 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 DeepSeek Search Everywhere MCP server expose?

DeepSeek Search Everywhere exposes 7 tools: deepseek_search, deepseek_fetch, deepseek_fetch_many, deepseek_research, deepseek_deep_research, and 2 more. Their descriptions and schemas cost roughly 1,885 tokens of context every time the server is loaded.

### Is the DeepSeek Search Everywhere MCP server still maintained?

DeepSeek Search Everywhere 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 DeepSeek Search Everywhere MCP server under?

DeepSeek Search Everywhere declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

### What did VerifyMCP test on the DeepSeek Search Everywhere MCP server?

VerifyMCP ran 16 checks on DeepSeek Search Everywhere, covering Malware scan, Known CVEs, Install scripts, Dependency health and 12 more. Endpoint Security and Transport & Reachability do not apply to a PyPI package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- PyPI project: https://pypi.org/project/deepseek-search-everywhere-mcp/
- Socket report: https://socket.dev/pypi/package/deepseek-search-everywhere-mcp
- Repository: https://github.com/radoxtech/deepseek-search-everywhere-mcp
- Changelog RSS feed: https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp.json
- HTML version of this page: https://verifymcp.io/servers/radoxtech-deepseek-search-everywhere-mcp/deepseek-search-everywhere-mcp