# StripFeed

> API proxy that converts any URL to clean, token-efficient Markdown for AI agents.

No LLM is used internally. Conversion is deterministic (Readability + Turndown). The `model` parameter is for cost tracking only.

## API

Base URL: https://www.stripfeed.dev

### Fetch Endpoint

GET /api/v1/fetch?url={encoded_url}
Authorization: Bearer sf_live_xxx

Returns clean Markdown from any URL. Strips ads, navigation, scripts, and noise.

Query parameters:
- url (required): Target URL (http/https)
- format: markdown (default), json, text, html
- selector: CSS selector to extract specific elements
- model: AI model ID for cost tracking (e.g. claude-sonnet-4-6, gpt-5)
- cache: Set to "false" to bypass cache
- ttl: Custom cache TTL in seconds (default 3600, max 86400)
- max_tokens: Truncate output to fit within a token budget

Response headers include token counts, savings percentage, cache status, and rate limit info.

### Usage Endpoint

GET /api/v1/usage
Authorization: Bearer sf_live_xxx

Returns current monthly usage, plan, limit, remaining requests, and reset date.

### Batch Endpoint

POST /api/v1/batch
Authorization: Bearer sf_live_xxx
Content-Type: application/json

Process up to 10 URLs in parallel in a single request.

Body: { "urls": ["https://news.ycombinator.com", {"url": "https://...", "selector": "article"}], "model": "claude-sonnet-4-6" }

## MCP Server

Install the StripFeed MCP server for Claude Code, Cursor, or any MCP-compatible client:

```
npx -y @stripfeed/mcp-server
```

Requires STRIPFEED_API_KEY env var. Exposes fetch_url and batch_fetch tools.

npm: https://www.npmjs.com/package/@stripfeed/mcp-server
GitHub: https://github.com/StripFeed/mcp-server

## TypeScript SDK

```
npm install stripfeed
```

```typescript
import StripFeed from "stripfeed";

const sf = new StripFeed("sf_live_your_api_key");
const result = await sf.fetch("https://docs.anthropic.com");
console.log(result.markdown);

// Just Markdown
const md = await sf.fetchMarkdown("https://docs.anthropic.com");

// Batch (Pro plan)
const batch = await sf.batch(["https://a.com", "https://b.com"]);
```

Zero dependencies. Node.js 18+. Full TypeScript types.

npm: https://www.npmjs.com/package/stripfeed
GitHub: https://github.com/StripFeed/stripfeed-js

## Python SDK

```
pip install stripfeed
```

```python
from stripfeed import StripFeed

sf = StripFeed("sf_live_your_api_key")
result = sf.fetch("https://docs.anthropic.com")
print(result["markdown"])

# Just Markdown
md = sf.fetch_markdown("https://docs.anthropic.com")

# Batch (Pro plan)
result = sf.batch(["https://a.com", "https://b.com"])
```

Zero dependencies. Python 3.9+. Type hints included.

PyPI: https://pypi.org/project/stripfeed/
GitHub: https://github.com/StripFeed/stripfeed-python

## x402 Pay-Per-Request (No API Key Needed)

GET /api/v2/fetch?url={encoded_url}

Same endpoint as v1, but uses the x402 payment protocol instead of API keys. AI agents pay $0.01 USDC per request on Base mainnet. No sign-up required.

How it works:
1. Send GET request to /api/v2/fetch?url=...
2. Server responds with 402 Payment Required + payment instructions
3. Client signs a USDC payment and resends with payment-signature header
4. Server verifies payment via facilitator, returns content, settles on-chain

Supports all query parameters: url, format, selector, cache, ttl, max_tokens.

Payment: $0.01 USDC on Base (eip155:8453) via x402 protocol.
Wallet: Set via X402_WALLET_ADDRESS env var.
Facilitator: https://x402.org/facilitator

Compatible clients: x402-cli, @x402/fetch, @x402/axios, any x402-compatible AI agent wallet (AgentCash, Coinbase CDP wallets).

## Pricing

- Free: 200 requests/month, 1 API key
- Pro: $29/month or $199/year, 100K requests/month, unlimited API keys
- Enterprise: Custom pricing, unlimited requests

## OpenClaw Skill

Install StripFeed as an OpenClaw skill from ClawHub:

```
clawhub install stripfeed
```

Set your API key in `~/.openclaw/openclaw.json`:

```json
{
  "skills": {
    "entries": {
      "stripfeed": {
        "enabled": true,
        "env": { "STRIPFEED_API_KEY": "sf_live_your_key" }
      }
    }
  }
}
```

ClawHub: https://clawhub.ai
GitHub: https://github.com/StripFeed/openclaw-skill

## AI Agent Discovery

- OpenAPI Schema: https://www.stripfeed.dev/api/v1/openapi
- AI Plugin Manifest: https://www.stripfeed.dev/.well-known/ai-plugin.json
- Agents.json: https://www.stripfeed.dev/.well-known/agents.json
- OpenClaw Skill: https://github.com/StripFeed/openclaw-skill
- Context7: https://context7.com/stripfeed/mcp-server

## Links

- Full API docs for LLMs: https://www.stripfeed.dev/llms-full.txt
- Docs: https://www.stripfeed.dev/docs
- MCP Server: https://www.npmjs.com/package/@stripfeed/mcp-server
- TypeScript SDK: https://www.npmjs.com/package/stripfeed
- Python SDK: https://pypi.org/project/stripfeed/
- Status: https://www.stripfeed.dev/status
- Support: https://www.stripfeed.dev/support