StripFeed

Convert any URL to clean Markdown optimized for LLM consumption. Strips ads, navigation, scripts, and noise. Returns token counts so you know exactly how much context you're using.

When to use this skill

Use StripFeed whenever you need to read a web page, article, documentation, or any URL content. It produces much cleaner output than raw HTML fetching and tells you the token cost.

Authentication

All requests require the STRIPFEED_API_KEY environment variable. Pass it as a Bearer token:

Authorization: Bearer $STRIPFEED_API_KEY

Get a free API key at https://www.stripfeed.dev (200 requests/month, no credit card).

Fetch a single URL

curl -s "https://www.stripfeed.dev/api/v1/fetch?url=URL_HERE" \
  -H "Authorization: Bearer $STRIPFEED_API_KEY"

This returns clean Markdown directly as the response body (Content-Type: text/markdown).

Parameters

JSON format (recommended for structured responses)

When you need metadata alongside the content, use format=json:

curl -s "https://www.stripfeed.dev/api/v1/fetch?url=URL_HERE&format=json" \
  -H "Authorization: Bearer $STRIPFEED_API_KEY"

JSON response includes:

{
  "markdown": "# Page Title\n\nClean content...",
  "url": "https://example.com",
  "title": "Page Title",
  "tokens": 1250,
  "originalTokens": 15000,
  "savingsPercent": 91.7,
  "cached": false,
  "fetchMs": 430,
  "format": "json",
  "truncated": false,
  "selector": null,
  "model": null
}

Response headers

Every response includes these headers:

• X-StripFeed-Tokens - Token count of the stripped content
• X-StripFeed-Original-Tokens - Token count of the raw HTML (before stripping)
• X-StripFeed-Savings-Percent - Percentage of tokens saved
• X-StripFeed-Cache - HIT or MISS
• X-StripFeed-Fetch-Ms - Time to fetch the URL (0 if cached)
• X-StripFeed-Format - Output format used

Batch fetch (Pro plan)

Fetch up to 10 URLs in a single request:

curl -s -X POST "https://www.stripfeed.dev/api/v1/batch" \
  -H "Authorization: Bearer $STRIPFEED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://example.com", "https://example.org"]}'

URLs can also include selectors:

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

Batch response:

{
  "results": [
    { "url": "...", "markdown": "...", "tokens": 1250, "status": 200 },
    { "url": "...", "error": "Failed to fetch", "status": 502 }
  ],
  "total": 2,
  "success": 1,
  "failed": 1
}

Individual URL errors don't break the batch. Check each result's status field.

Error handling

Tips

• Default format is markdown which returns raw text. Use format=json when you need token counts and metadata in the response body.
• Responses are cached for 1 hour by default. Use cache=false for real-time content.
• The max_tokens parameter is useful to fit content within your context window budget.
• Use selector to grab only the main content (e.g. selector=article or selector=.post-content) and skip sidebars/footers.
• Free plan: 200 requests/month, markdown format only. Pro plan ($19/mo): 100K requests, all formats, selectors, batch endpoint.