← All agent tools

Web search Parallel Web Systems

Parallel MCP server (free, no key)

Bring web search, page extraction and evidence oriented answers into an agent through Parallel’s public MCP connection.

Source review: 2026-10-28 · Imported: 1 November 2026

A practical starting point

What could you build?

A comparison assistant can gather web evidence before producing a structured answer.

Before you connect

Do not assume an unlimited free allowance when a numeric limit is unpublished. Keep source links with any generated answer.

Capabilities & requirements

Dated specifications from the licensed source dataset. “Not published” indicates missing information, rather than confirmed absence.

Parallel MCP server (free, no key) specifications
CapabilityDetails and source
StatusNo MCP label; decisions tool early

Search and Extract are GA at /v1; the docs call web_decisions an early release.

Source (external) · Verified 2026-10-28 on https://docs.parallel.ai/integrations/mcp/quickstart.md
API key requiredNo key for 3 tools

web_search, web_fetch and web_decisions answered without auth on 2026-10-28; web_research needs OAuth or a key.

Source (external) · Verified 2026-10-28 by test request to https://docs.parallel.ai/integrations/mcp/quickstart.md
Free tierFree without a key

Keyless calls are free and report no usage; keyed calls bill at API rates (Search fast $1 per 1,000).

Source (external) · Verified 2026-10-28 on https://docs.parallel.ai/integrations/mcp/quickstart.md
Rate limitNot published

Keyless limit not published (“lower rate limits”); our second keyless search within about a minute got HTTP 429. Paid API default 600/min.

Source (external) · Not published by vendor: checked 2026-10-28 on https://docs.parallel.ai/getting-started/rate-limits.md
Official MCP serverHosted, mcp.parallel.ai/mcp

Streamable HTTP; tools/list run 2026-10-28 returned web_search, web_fetch and web_decisions.

Source (external) · Verified 2026-10-28 by test request to https://docs.parallel.ai/integrations/mcp/quickstart.md
Official SDKPython + TypeScript

parallel-web on PyPI (1.3.7) and npm (1.3.6), for the underlying APIs.

Source (external) · Verified 2026-10-28 on https://docs.parallel.ai/llms.txt
Data license / termsParallel Customer Terms

Effective Oct 17, 2026; no model training on outputs, no data reselling; Parallel may use inputs to improve services.

Source (external) · Verified 2026-10-28 on https://parallel.ai/customer-terms
Commercial use allowedYes, for your business use

Allowed for the customer’s business purposes, with the training, reselling and competitive-use bans in section 2.

Source (external) · Verified 2026-10-28 on https://parallel.ai/customer-terms
CORS / browser useAccess-Control-Allow-Origin: *

Returned on our MCP POST on 2026-10-28, with Mcp-Session-Id exposed.

Source (external) · Verified 2026-10-28 by test request to https://docs.parallel.ai/integrations/mcp/quickstart.md
Freshness / coverageLive web; cached pages allowed

web_fetch fetches live by default (allow_live_fetch true); cached content can be older.

Source (external) · Verified 2026-10-28 on https://docs.parallel.ai/integrations/mcp/quickstart.md

Operational reference

Additional facts and qualifications retained from AgentAPIList’s CC BY 4.0 dataset, with provider references.

Price

Keyless calls are free and report no usage. With OAuth or an API key, calls bill at API rates: Search $1 per 1,000 requests in the fast mode the MCP uses by default, Extract $1 per 1,000 URLs, Decisions $1 per 1,000 requests.

Provider reference (external)

API key / auth

None for web_search, web_fetch and web_decisions. OAuth (connect to ?login) or a Bearer or x-api-key header unlocks web_research and higher limits. Our keyless initialize, tools/list and web_search calls on 2026-10-28 returned HTTP 200.

Provider reference (external)

Rate limit

The keyless tier is “free at lower rate limits” with no number published. In the source publisher’s test a second keyless search about a minute after the first returned HTTP 429 with a “free-tier rate limit” message. Paid API defaults are 600 requests a minute for Search, Extract and Decisions.

Provider reference (external)

License / terms

Parallel’s Customer Terms (effective Oct 17, 2026), linked from the server’s x-parallel-terms header, apply. They ban using outputs to train models or to build data-reselling products, and give Parallel a license to use your inputs to improve its services.

Provider reference (external)

Version status

No stage label on the MCP server itself. Search and Extract are GA at /v1; the docs call web_decisions an early release.

Provider reference (external)

Tools

Keyless: web_search, web_fetch (up to 20 URLs per call), web_decisions (1 to 100 typed questions). With auth: web_research. Our tools/list returned the three keyless tools. One older docs page lists only web_search and web_fetch for keyless use.

Provider reference (external)

Setup references

HTTP request example

Adapted from the attributed source dataset. Review before running; these examples were not executed by Tech in General.

# 1) start a session (keyless)
curl -sS -D - https://mcp.parallel.ai/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"test","version":"1.0"}}}'

# 2) search, reusing the mcp-session-id header from step 1
curl -sS https://mcp.parallel.ai/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "mcp-session-id: $SID" \
  -d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"web_search","arguments":{"objective":"Find the official NPPES NPI Registry API documentation page","search_queries":["NPPES NPI Registry API","NPI registry API version 2.1"]}}}'
Agent connection example

Adapted from the attributed source dataset. Review before running; these examples were not executed by Tech in General.

# Claude Code (keyless)
claude mcp add --transport http parallel https://mcp.parallel.ai/mcp
# Codex
codex mcp add parallel --url https://mcp.parallel.ai/mcp

# Cursor: .cursor/mcp.json
{
  "mcpServers": {
    "parallel": { "url": "https://mcp.parallel.ai/mcp" }
  }
}
# Only the tools you want: https://mcp.parallel.ai/mcp?tools=web_search,web_fetch
# Higher limits + web_research: 'https://mcp.parallel.ai/mcp?login' (OAuth) or an Authorization: Bearer header

A tool schema needs an HTTP implementation; it does not by itself install an MCP server. Keep credentials out of browser code.

Other connections to explore

Back to the full directory →

Source & reuse

Download the complete licensed source snapshot (JSON), including the original source’s caveats, alternative notes and examples.