← All agent tools

Web search TinyFish

TinyFish Search & Fetch (free tier)

Combine web discovery with browser rendered page extraction using separate Search and Fetch endpoints.

Source review: 2026-10-18 · Imported: 1 November 2026

A practical starting point

What could you build?

Find relevant pages first, then retrieve a readable version of the few pages needed for a comparison.

Before you connect

Create an account and keep the API key on a server. Search and fetch allowances are tracked separately.

Capabilities & requirements

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

TinyFish Search & Fetch (free tier) specifications
CapabilityDetails and source
StatusNot published

The docs carry no beta or preview label for Search and Fetch; only Enterprise plans list an uptime SLA.

Source (external) · Not published by vendor: checked 2026-10-18 on https://docs.tinyfish.ai/
API key requiredX-API-Key (free account)

Our request without a key returned HTTP 401 MISSING_API_KEY.

Source (external) · Verified 2026-10-18 by test request to https://api.search.tinyfish.ai?query=sec+edgar+api
Free tier12,000 searches + 1,000 fetches/day

Works at a $0 wallet balance with no card; beyond the daily allowance requests need wallet funds.

Source (external) · Verified 2026-10-18 on https://docs.tinyfish.ai/search-api
Rate limitSearch 30/min, Fetch 150 URLs/min

Free-tier limits from TinyFish’s May 4, 2026 announcement; Fetch takes up to 10 URLs per call.

Source (external) · Verified 2026-10-18 on https://www.tinyfish.ai/blog/search-and-fetch-are-now-free-for-every-agent-everywhere
Official MCP serverHosted, agent.tinyfish.ai/mcp

OAuth sign-in; also exposes the paid Agent and Browser tools.

Source (external) · Verified 2026-10-18 on https://docs.tinyfish.ai/mcp-integration/index.md
Official SDKPython and TypeScript

Both read TINYFISH_API_KEY from the environment; a CLI is also published.

Source (external) · Verified 2026-10-18 on https://docs.tinyfish.ai/authentication
Data license / termsTinyFish Terms of Service

Internal business or personal use; no resale or sublicensing; TinyFish may use service data to train its models.

Source (external) · Verified 2026-10-18 on https://www.tinyfish.ai/terms
Commercial use allowedInternal business use

Licensed for your internal business purposes; reselling and building a competing service are not allowed.

Source (external) · Verified 2026-10-18 on https://www.tinyfish.ai/terms
CORS / browser useAccess-Control-Allow-Origin: *

Preflight to both endpoints returned 200 with *; a browser app would expose your key, so proxy it.

Source (external) · Verified 2026-10-18 by test request to https://api.fetch.tinyfish.ai
Freshness / coverageLive fetch; cache unless ttl set

Fetch may serve a cached copy by default; ttl: 0 forces a live fetch.

Source (external) · Verified 2026-10-18 on https://docs.tinyfish.ai/fetch-api

Operational reference

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

Price

Search is free up to 12,000 requests a day and Fetch up to 1,000 URLs a day. Above that, requests need a positive wallet balance. The pricing page says Search and Fetch run at any wallet balance with no card required.

Provider reference (external)

API key / auth

Every REST request needs an X-API-Key header with a key from agent.tinyfish.ai/api-keys (free account). Our request without a key on 2026-10-18 returned HTTP 401 MISSING_API_KEY. The hosted MCP server at agent.tinyfish.ai/mcp signs in with OAuth instead.

Provider reference (external)

Rate limit

Free tier: 30 Search requests per minute and 150 Fetch URLs per minute (TinyFish announcement, May 4, 2026), on top of the daily allowances above. Fetch takes up to 10 URLs per request, and failed URLs do not count against the quota.

Provider reference (external)

Terms

TinyFish Terms of Service (effective Aug 17, 2026): use is licensed for your internal business or personal purposes; no reselling or sublicensing the service, and no using it or its outputs to build a competing service. The terms also let TinyFish use service data to train its own models, with safeguards.

Provider reference (external)

Fetch freshness

Fetch may return a cached copy by default. Send ttl: 0 to force a live fetch, or a number of seconds to accept cached entries younger than that.

Provider reference (external)

Status

The docs carry no beta or preview label for Search and Fetch. Enterprise plans add an uptime SLA; the free tier has none published.

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.

# Not run: needs a TinyFish API key (free account at agent.tinyfish.ai/api-keys)
curl "https://api.search.tinyfish.ai?query=sec+edgar+xbrl+api" \
  -H "X-API-Key: $TINYFISH_API_KEY"

curl -X POST https://api.fetch.tinyfish.ai \
  -H "X-API-Key: $TINYFISH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://www.sec.gov/about/developer-resources"], "format": "markdown"}'
Agent connection example

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

// Hosted MCP (OAuth sign-in on first use)
// Claude Code:
//   npx -y install-mcp@latest https://agent.tinyfish.ai/mcp --client claude-code
// Codex:
//   codex mcp add tinyfish --url https://agent.tinyfish.ai/mcp
// Cursor .cursor/mcp.json
{
  "mcpServers": {
    "tinyfish": { "url": "https://agent.tinyfish.ai/mcp" }
  }
}

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.