We and selected third parties use cookies or similar technologies for technical purposes and, with your consent, to analyze how the site is used.

You can freely give, refuse or withdraw your consent at any time.

Parallel Search

MCP server by Parallel Web Systems

Search the live web and extract page content, with excerpts formatted for AI agents to answer from directly.

No sign-in (No account or key needed)Streamable HTTP

No sign-in needed. Connect with Parallel Search’s own URL:

https://search.parallel.ai/mcp

Paste it into any client that supports remote MCP servers. Your client talks to Parallel Web Systems directly, not through mcp.tc.

Share this server

Opens this page, with the URL and setup steps.

mcp.tc/i/parallel-search

Connects an assistant to Parallel's web search and page extraction. web_search returns excerpts meant to answer questions directly, and accepts several search queries in one call. web_fetch pulls relevant content from specific URLs when the excerpts are not enough.

Runs as a hosted streamable HTTP endpoint at https://search.parallel.ai/mcp and works without authentication. An OAuth variant is available at https://search.parallel.ai/mcp-oauth. Use is subject to the Parallel Customer Terms and Privacy Policy.

  • Search the web for current information
  • Run multiple search queries in one call
  • Get excerpts that can answer questions directly
  • Fetch and extract content from specific URLs
  • Focus extraction on a stated objective
  • Retrieve full page content when needed
  • web_searchRead-only

    Web search

  • web_fetchRead-only

    Web fetch

  • send_feedbackWrites

    Send feedback to Parallel

Read from the live server on 17 Nov 2026. Read-only, Writes and Can delete are hints the server declares; your client decides whether to ask before running a tool.

  • Search for what changed in the EU AI Act this month.

  • Find the latest funding rounds for AI infrastructure startups.

  • Fetch this docs page and quote the section on rate limits.

  • Compare the pricing pages of these three vendors.

Every client connects to search.parallel.ai. Pick yours; the page remembers your choice.

Claude Code

  1. Run this in a terminal, in your project folder:
claude mcp add --transport http parallel-search https://search.parallel.ai/mcp
  1. Start Claude Code and type /mcp. parallel-search should show as connected.

Add --scope user to make it available in every project, not just this one.

Claude Desktop

  1. Open Customize → Connectors, click + Add, then Add custom connector.
  2. Name it Parallel Search and paste this URL:
https://search.parallel.ai/mcp
  1. Click Add. Its tools appear in the chat’s tools menu.

Claude Desktop’s JSON config file only starts local servers. Remote servers go through Connectors, and connectors you add on claude.ai show up here too.

claude.ai

  1. Open the connector form on claude.ai. This button fills in the name and URL for you:
  1. Check that the URL reads https://search.parallel.ai/mcp and click Add.
  2. Turn it on in a chat from the tools menu.

Free plans allow one custom connector. On Team and Enterprise plans an owner adds it under Organization settings → Connectors.

ChatGPT

  1. On chatgpt.com, go to Plugins, select +, then Add custom MCP server.
  2. Enter a name for the server.
  3. Under Connection, paste https://search.parallel.ai/mcp as the server URL, then for authentication choose No authentication.
  4. Check I understand and want to continue, then select Create as a plugin.

Whether you can add a custom MCP server depends on your ChatGPT plan and your workspace permissions (in Business workspaces, admins and owners can).

Cursor

Or add it by hand to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (this project):

mcp.json
{
  "mcpServers": {
    "parallel-search": {
      "url": "https://search.parallel.ai/mcp"
    }
  }
}

VS Code

Or from a terminal:

code --add-mcp '{"name":"parallel-search","type":"http","url":"https://search.parallel.ai/mcp"}'

Or commit it to the repo in .vscode/mcp.json:

.vscode/mcp.json
{
  "servers": {
    "parallel-search": {
      "type": "http",
      "url": "https://search.parallel.ai/mcp"
    }
  }
}

Devin Desktop

  1. Add it to ~/.config/devin/mcp_config.json (macOS and Linux) or %APPDATA%\devin\mcp_config.json (Windows):
mcp_config.json
{
  "mcpServers": {
    "parallel-search": {
      "serverUrl": "https://search.parallel.ai/mcp"
    }
  }
}
  1. Refresh the MCP server list in Cascade.

Devin Desktop is the new name for Windsurf. It reads serverUrl (or url) for remote servers.

Codex

codex mcp add parallel-search --url https://search.parallel.ai/mcp

Or edit ~/.codex/config.toml directly:

config.toml
[mcp_servers.parallel-search]
url = "https://search.parallel.ai/mcp"

Gemini CLI

gemini mcp add --transport http parallel-search https://search.parallel.ai/mcp

This adds it to the current project. Add -s user to use it everywhere.

Any client

Most clients accept this shape. Some name the URL field differently: serverUrl in Devin Desktop, httpUrl in Gemini CLI’s settings file.

{
  "mcpServers": {
    "parallel-search": {
      "type": "http",
      "url": "https://search.parallel.ai/mcp"
    }
  }
}

Zed puts servers under context_servers in its settings. Cline needs "type": "streamableHttp", or it assumes SSE.

Client only starts local servers? Bridge it with npx -y mcp-remote https://search.parallel.ai/mcp.

Show Parallel Search in a README or on your site. Each embed links to this page, uses the server’s own URL, and sets no cookies. How embeds work

The card, the iframe and the JSON are in English. Add ?lang=it, fr, de or es to their address for another language.

We read this server’s tool list every 3 hours and record what changes. Last read 18 Nov 2026, 05:35 UTC.

Follow Parallel Search and we’ll email you when something changes.

  1. , Automatic check

    2 tools changed.

    Details, 17 Nov 2026, 20:55 UTC

    Changed 2

    • web_fetchoutput schema
    • web_searchoutput schema
  2. , Automatic check

    1 tool added. Server instructions changed.

    New server instructions

    Details, 7 Nov 2026, 02:45 UTC

    Added 1

    • send_feedbackWrites

      Report a problem with Parallel to the Parallel team. Use it when, while working with Parallel, you see any of these: - A Parallel tool returned an error you didn't cause: a 5xx, a timeout, or a rate limit you didn't expect. - Results were wrong, irrelevant, stale or missing something you know exists: a wrong date or number, an excerpt the page doesn't support, an empty fetch of a page that has content. - Parallel's docs, tool descriptions and behavior disagree, or you couldn't work out how to call a tool. - An evaluation, benchmark or comparison you ran shows Parallel doing worse than expected. Problems you worked around count too. It isn't for mistakes in your own requests, or for questions. Ask the user first. When you tell them about the problem, offer to report it and show what you'd send: the message and the IDs. Send it only if they agree, then tell them it was sent. Put the IDs involved (search_id, extract_id, run_id) in `examples`, with a short note on each. web_decisions retu…

    Server instructions

    Before

    The search and fetch tools are read-only, idempotent, and safe to call concurrently. When several independent search or fetch objectives are already known, issue one tool call per objective concurrently; do not wait for one result before starting another. Batch related search queries or URLs that share a common objective into one call. Use web_search first for most factual, current-information, research, comparison, documentation, and troubleshooting questions. The web_search results include excerpts intended to be useful for answering directly. If the excerpts contain enough evidence, answer from them — do not fetch every search result by default. For broad tasks you can issue multiple `search_queries` in a single web_search call rather than chaining calls. Use web_fetch when the user asks about a specific URL or page, needs exact wording or quotes, requests full-page analysis, or when search excerpts are conflicting or clearly insufficient. Use web_fetch only when web_search excerpts are not enough: the user asked about a specific URL or page, you need exact wording or quotes, you need full-page analysis, or the excerpts are conflicting or clearly insufficient. Do not fetch every search result by default — search excerpts are usually answer-ready. When you do fetch, you can pass multiple related URLs in one call along with a common `objective`; the tool will return the parts most relevant to it. Use of this server is subject to the Parallel Customer Terms (https://parallel.ai/customer-terms) and Privacy Policy (https://parallel.ai/privacy-policy).

    After

    If a Parallel tool fails, rate-limits you, contradicts its docs, or returns empty, truncated, wrong or stale results, tell the user and offer to report it to Parallel with send_feedback, citing the request IDs. Send only if they agree. The search and fetch tools are read-only, idempotent, and safe to call concurrently. When several independent search or fetch objectives are already known, issue one tool call per objective concurrently; do not wait for one result before starting another. Batch related search queries or URLs that share a common objective into one call. Use web_search first for most factual, current-information, research, comparison, documentation, and troubleshooting questions. The web_search results include excerpts intended to be useful for answering directly. If the excerpts contain enough evidence, answer from them — do not fetch every search result by default. For broad tasks you can issue multiple `search_queries` in a single web_search call rather than chaining calls. Use web_fetch when the user asks about a specific URL or page, needs exact wording or quotes, requests full-page analysis, or when search excerpts are conflicting or clearly insufficient. Use web_fetch only when web_search excerpts are not enough: the user asked about a specific URL or page, you need exact wording or quotes, you need full-page analysis, or the excerpts are conflicting or clearly insufficient. Do not fetch every search result by default — search excerpts are usually answer-ready. When you do fetch, you can pass multiple related URLs in one call along with a common `objective`; the tool will return the parts most relevant to it. Use of this server is subject to the Parallel Customer Terms (https://parallel.ai/customer-terms) and Privacy Policy (https://parallel.ai/privacy-policy).

  3. , Automatic check

    1 tool changed.

    Details, 27 Oct 2026, 21:03 UTC

    Changed 1

    • web_fetchoutput schema
  4. , Automatic check

    1 tool changed.

    Details, 24 Oct 2026, 16:36 UTC

    Changed 1

    • web_searchinput schema
      • Description of search_queries
        Before

        Concise keyword search queries, 3-6 words each, which may include search operators. At least one query is required; provide 2-3 for best results. For broad tasks, you can include multiple related que…

        After

        Concise keyword search queries, 3-6 words each. They may use site: with a bare domain (such as site:example.com), but never a URL path after the domain and never quoted phrases. At least one query is…

  5. , Automatic check

    1 tool changed.

    Details, 22 Oct 2026, 19:39 UTC

    Changed 1

    • web_fetchoutput schema
  6. , Automatic check

    Tracking started: 2 tools.

Can I paste mcp.tc/i/parallel-search into my MCP client?

No. mcp.tc links are pages, not server addresses. Connect with https://search.parallel.ai/mcp, so your client talks to Parallel Search directly. The quick link is for sharing: it opens this page, with setup steps for every client.

Does Parallel Search need an API key or a sign-in?

No. Parallel Search doesn’t ask for an account or key.

Is Parallel Search a remote or a local server?

Remote. Parallel Web Systems hosts it at https://search.parallel.ai/mcp, and it speaks Streamable HTTP. There’s nothing to install.

What can Parallel Search do?

It has 3 tools: web_search, web_fetch and send_feedback. You can search the web for current information, run multiple search queries in one call and get excerpts that can answer questions directly.

Which clients can use it?

Any client that supports remote MCP servers: Claude Code, Claude Desktop, claude.ai, ChatGPT, Cursor, VS Code, Devin Desktop, Codex, Gemini CLI, Zed and others. The setup steps cover each one.

Does mcp.tc track changes to Parallel Search?

Yes. Parallel Search’s History tab shows each change we find to its tools, down to descriptions and parameters, plus new versions and edits by its owner. Follow the listing to get an email when it changes.

Who wrote this page?

mcp.tc’s robot read Parallel Search’s own metadata (its MCP handshake, tool list and public pages), and an AI model drafted the text from it. A person reviews anything the checks can’t confirm. The text can still be wrong, so if you spot a mistake, use Report this listing on this page.

mcp.tc isn’t affiliated with Parallel Web Systems. This page was built from Parallel Search’s public metadata, last checked on 17 Nov 2026, and an AI model wrote the description, so it can be wrong. Names and marks belong to their owners. Something off? Report this listing.