Skip to content
Agentmart
Submit a tool

mcp-retriever

One MCP server for web search, page extraction, and cited research.

Connector by keiranhaax · GitHub account verified by the MCP Registry

IndexedMITLocal (stdio)★ 1
Write a review

Chalk Labs

Need help deploying mcp-retriever?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/keiranhaax-mcp-retriever

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

One MCP server for web search, page extraction, and cited research.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 26, 2026.

Tools
17 read · 1 write · 0 delete
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Tools by what they do 18

Read from the source code; the running server may list different tools.

Read 17 · Look things up without changing anything.

  • ai_search
  • brave_llm_context
  • brave_media_search
  • brave_news_search
  • context_brand_intel
  • context_classify
  • context_news
  • context_styleguide
  • context_transaction_identify
  • context_web_extract
  • firecrawl_agent
  • github_search
Show 5 more tools
  • result_read
  • search_and_read
  • web_read
  • web_search
  • web_search_fused

Write 1 · Create, change, send or run things.

  • web_extract

Network hosts it talks to 29

  • api-dashboard.search.brave.com · found in code
  • api.context.dev · found in code
  • api.exa.ai · found in code
  • api.firecrawl.dev · found in code
  • api.github.com · found in code
  • api.linkup.so · found in code
  • api.search.brave.com · found in code
  • api.tavily.com · found in code
  • api.you.com · found in code
  • app.linkup.so · found in code
  • app.tavily.com · found in code
  • archive.org · found in code
  • brave.com · found in code
  • context.dev · found in code
  • dashboard.exa.ai · found in code
  • docs.exa.ai · found in code
  • docs.firecrawl.dev · found in code
  • docs.searxng.org · found in code
  • docs.tavily.com · found in code
  • example.test · found in code
  • firecrawl.dev · found in code
  • github.com · found in code
  • guard.local · found in code
  • iana.org · found in code
  • keiranhaax.dev · found in code
  • raw.githubusercontent.com · found in code
  • search.brave.com · found in code
  • unpkg.com · found in code
  • you.com · found in code

Files it touches

  • /home/ubuntu/.pm2/logs/mcp-retriever-error.log
  • /home/ubuntu/.pm2/logs/mcp-retriever-out.log
  • /home/ubuntu/backups/mcp-omnisearch-p3a-20260907-EFIkoK/RESULT.md
  • /home/ubuntu/worktrees/mcp-omnisearch-evolution-p0-20260906
  • /opt/mcp-retriever
  • /opt/mcp-retriever/start-server.sh
  • /tmp/fixture-results
  • /tmp/omnisearch-evolution-p1b-HF26M2ho
  • /tmp/omnisearch-evolution-p2-PsHVP7aD
  • /tmp/omnisearch-p3a-gate-yNfmVZ
  • Your home directory (~)

Tools (18)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • ai_searchread-only

    Description is built at runtime.

  • brave_llm_contextread-only

    Research Context/RAG: retrieve LLM-ready grounding chunks from Brave. Use for dense context, not ordinary link search.

  • brave_media_searchread-only

    Media Search: find Brave image or video results with SafeSearch and country/language filters. Use only when visual media results are needed.

  • brave_news_searchread-only

    News Search: find recent news articles with freshness, pagination, country/language, SafeSearch, and snippets.

  • context_brand_intelread-only

    Context.dev Brand: retrieve brand/company intelligence such as logos, colors, socials, descriptions, industries, and domain/company/ticker matches.

  • context_classifyread-only

    Context.dev Business: classify companies by NAICS, SIC, or EIC using domain or company name.

  • context_newsread-only

    Context.dev News: search news by natural-language query, or get company-keyed news by name, domain, ticker, or ISIN. Filter by article type, publisher, and date window; page with cursor.

  • context_styleguideread-only

    Context.dev Design: extract styleguide signals and optional fonts for a domain or direct URL.

  • context_transaction_identifyread-only

    Context.dev Transactions: identify messy bank/card transaction descriptors as real brands or companies.

  • context_web_extractread-only

    Context.dev Web: scrape markdown/HTML/images/screenshots, crawl pages, get sitemaps, map site URLs by topic (mode=map), or web-search (mode=web_search; limit sets result count) with optional query-relevant highlights, language filter, and markdown.

  • firecrawl_agentread-only

    Autonomous Web Agent: Credit-sensitive; start a web task (default cap 100 credits), then use action=status or cancel with job_id. Repeating start creates a new paid job. Prefer search/extract for simple tasks.

  • github_searchread-only

    Search GitHub: find code, repositories, or users. Supports filename:, path:, repo:, user:, language:, and in:file syntax.

  • result_readread-only

    Description is built at runtime.

  • search_and_readread-only

    Search and read: retrieve top-ranked public pages with explicit providers, bounded requests and per-source evidence. No synthesis, crawling or fallback. Use separate search/extract when selecting sources needs judgment.

  • web_extract

    Extract/Process: read or process known URLs. Firecrawl handles scrape/summarize/crawl/map/extract/actions/search; Exa handles contents/similar; Tavily extracts.

Show 3 more tools
  • web_readread-only

    Read 1-20 public URLs with basic Tavily extraction, Exa contents, Firecrawl scraping, or Context.dev query highlights (query required). Returns source evidence, not synthesis. No crawl, actions, or advanced provider options.

  • web_searchread-only

    Search: find web pages, articles, or data. Use Brave for operators, Exa for semantic or deep search, Tavily for factual results, and You.com as a fast fallback.

  • web_search_fusedread-only

    Search 2-3 explicit web providers in parallel and merge their results with reciprocal rank fusion (k=60) after URL canonicalisation. Reports each provider's outcome and cost; partial results when some fail. No synthesis or fallback.

Signals

Stars
1
Forks
0
Open issues
0
Last commit
Oct 20, 2026 (14 days ago)
Last release
Oct 8, 2026 (26 days ago)

From GitHub (fetched Oct 24, 2026).

Details

Registry name
io.github.keiranhaax/mcp-retriever
Version
0.0.24
Auth
No auth
Packages
npm mcp-retriever
License
MIT
Repository
github.com/keiranhaax/mcp-retriever
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
Official MCP Registrygithub-graphql

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get mcp-retriever running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying mcp-retriever?

CallGet help
Cite this page

mcp-retriever. Agent Mart. https://agentmart.dev/connectors/keiranhaax-mcp-retriever (last updated 2026-10-24)