Retrieval MCP Server

An MCP server that gives an LLM three web tools: search, image search, and page scraping - no API keys required. Its README documents 3 tools, including web_search, web_search_images, web_scrape.

People who work with bing images, duckduckgo and image search and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Go.

VERIFIED ACTIVE

LAST COMMIT 2026-09-03 · ★ 12 · #42 OF 51 MAINTAINED SEARCH · VERIFIED 2026-09-18

MIT · Go servers · how we verify → /methodology

01 · Install Retrieval

before you install - you'll need

Set MAX_RESULTS, DEFAULT_TIMEOUT_MS before connecting.

Docker

docker run -i --rm ghcr.io/role1776/mcp-retrieval:0.1.7

Claude Desktop - add to config

{
  "mcpServers": {
    "role1776-mcp-retrieval": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/role1776/mcp-retrieval:0.1.7"
      ]
    }
  }
}
Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as container (OCI image) - can be sandboxed away from your filesystem · Claude Desktop extension (.mcpb bundle)

repo age created 2026-08-11 - young repo, little track record yet

license MIT - declared in the repository

registry namespace io.github.Role1776 is GitHub-verified and matches the repo owner

03 · What Retrieval can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

What you can build

With this server connected, an agent can run one or more queries in parallel and return per-query deduplicated, reranked snippets with links, run one or more image queries in parallel and return per-query deduplicated image results, and download one or more pages in parallel and return the main article text as Markdown.

The 3 tools it gives your agent

Extracted from the project's README - what MCP Retrieval lets an agent do.

web_search
- Runs one or more queries in parallel and returns per-query deduplicated, reranked snippets with links.
web_search_images
- Runs one or more image queries in parallel and returns per-query deduplicated image results.
web_scrape
- Downloads one or more pages in parallel and returns the main article text as Markdown.

Latest releases

v0.1.7 · 2026-08-17

relax Image validation rules and handle ErrNoRelevantImages as a non-fatal result state (@Role1776) · enforce English language for web and image search queries and fix scrape error handling in tests (@Role1776)

v0.1.6 · 2026-08-17

add URL validation constraints to ImageProps and update tests to include validation cases (@Role1776) · encapsulate domain entities by unexporting fields and introducing view-based mapping to DTOs (@Role1776)

v0.1.5 · 2026-08-15

remove cross-field validation for search limits to allow flexible configuration and add unit tests for limit logic in usecase (@Role1776)

04 · Who maintains Retrieval

MCP Retrieval is maintained by role1776. We track 2 MCP servers from role1776 - 2 actively maintained, 13 combined GitHub stars, oldest repo from Aug 2026.

05 · Facts

category
search - ranked #42 of 51 actively-maintained search servers as of 2026-09-18.
release cadence
8 releases in the last 90 days (latest 2026-08-17)
registry
io.github.Role1776/mcp-retrieval (active, first published 2026-08-11 · 8 versions)
packages
oci:ghcr.io/role1776/mcp-retrieval:0.1.7 · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_darwin_amd64.mcpb · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_darwin_arm64.mcpb · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_linux_amd64.mcpb · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_linux_arm64.mcpb · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_windows_amd64.mcpb · mcpb:https://github.com/Role1776/mcp-retrieval/releases/download/v0.1.7/mcp-retrieval_0.1.7_windows_arm64.mcpb

06 · Retrieval FAQ

Is Retrieval still maintained?

Yes - as of 2026-09-18, its last commit was 2026-09-03 and it shipped 8 releases in the last 90 days. We re-verify nightly.

What can Retrieval do?

With this server connected, an agent can run one or more queries in parallel and return per-query deduplicated, reranked snippets with links, run one or more image queries in parallel and return per-query deduplicated image results, and download one or more pages in parallel and return the main article text as Markdown.

How do I install Retrieval?

Run `docker run -i --rm ghcr.io/role1776/mcp-retrieval:0.1.7`. The README documents 2 environment variables (MAX_RESULTS, DEFAULT_TIMEOUT_MS) to set first. Set MAX_RESULTS, DEFAULT_TIMEOUT_MS before connecting. You can also paste the ready-made client config above.

07 · Alternatives to Retrieval