TinyFish Fetch turns any URL into clean markdown, html, or JSON for grounded agent workflows.
# Rate limits
Limits apply per API key, per minute.
| Tier | Requests/min | Burst |
| ---------- | ------------ | ------ |
| Pro | 600 | 1,000 |
| Enterprise | 6,000 | 10,000 |
6,000 requests a minute on enterprise, bursting to 10,000.
A URL in. Clean, structured page context out.
Trusted by
TinyFish Fetch loads pages in a real browser before extraction, then returns the clean page context agents need. JavaScript-rendered content and dynamic layouts come through without making teams manage browser infrastructure.
Navigation, cookie banners, ads, repeated footers, and boilerplate fall away. Relevant text, headings, lists, and document structure stay, giving agents cleaner input for better reasoning.
Fetch returns clean markdown by default, with HTML and JSON options when your workflow needs structure for retrieval, monitoring, or storage.
Fetch gives teams control over freshness, from fast cached reads to live page context when the workflow needs the latest version.
Call Fetch from the CLI, MCP, SDK, or API. Same clean page context on every surface.
TypeScript SDK
Skills
LangChain
REST API
Codex
Hermes Agent
n8n
MCP
Cursor
OpenClaw
Dify
Python SDK
Claude Code
CLITypeScript SDKSkillsLangChainREST APICodexHermes Agentn8nMCPCursorOpenClawDifyPython SDKClaude CodeCLIREST APIPython SDKOpenClawCodexClaude CodeSkillsCursorTypeScript SDKHermes AgentCLIn8nLangChainMCPDifyREST APIPython SDKOpenClawCodexClaude CodeSkillsCursorTypeScript SDKHermes AgentCLIn8nLangChainMCPDifyHermes AgentMCPTypeScript SDKSkillsREST APICodexLangChainCLIn8nClaude CodePython SDKCursorDifyOpenClawHermes AgentMCPTypeScript SDKSkillsREST APICodexLangChainCLIn8nClaude CodePython SDKCursorDifyOpenClawFetch quality is measured by whether agents get usable context from real URLs: broad coverage and high signal in the returned content.
93%
of pages came back as usable context — 42 of the 45 tested
Coverage · 45 fixed URLs · internal evaluation
90.3%
of what comes back is signal, not navigation, ads and boilerplate
Median signal ratio · 45 fixed URLs · internal evaluation
$0
Many scraping APIs meter markdown extraction by credit or page. TinyFish Fetch is free on every plan.
TinyFish Fetch is a web scraping API for programmatic workflows that need clean page context from live webpages. It returns markdown, HTML, or JSON, so agents and apps can use page content in grounded workflows.
Fetch can return markdown, HTML, or JSON. Markdown is the default for clean LLM-readable content, HTML preserves semantic markup, and JSON returns a structured document tree.
TinyFish Fetch is built for agent workflows that need grounded page context, not raw scrape output. It renders pages when needed, extracts clean context for models, and gives developers TTL controls for choosing cached or live reads.
Yes. Fetch is free on every TinyFish plan. No credit card is required to start.
Use Fetch when your workflow already has a URL and needs grounded page context for research, retrieval, monitoring, citations, or downstream analysis.
Use Fetch to turn live webpages into clean, complete context for research, retrieval, monitoring, and grounded decision workflows.
No credit card required
