# Semantic HTML Pruner with BM25 Budget Control

> Semantic HTML Pruner with BM25 Budget Control is a paid API for AI agents from llms-generator.agentdiet.dev, paid per call via MPP, $0.001/call, status unknown (last checked 2026-10-23).

Fetches a URL or accepts raw HTML and distills it into token-efficient Markdown, structured JSON, skeleton outline, or BM25-ranked budget excerpt for AI agent consumption

## Facts

- Endpoint: POST https://llms-generator.agentdiet.dev/v1/semantic-html-pruner?utm_source=zero.xyz
- Price: $0.001/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-23
- Activations on Zero: 0
- Provider: llms-generator.agentdiet.dev
- Website: https://llms-generator.agentdiet.dev
- Canonical page: https://www.zero.xyz/c/llms-generator-agentdiet-dev-semantic-html-pruner-with-bm25-budget-8d7fc6e8
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_xljebk4N8cfCK7Nsm53XN

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability llms-generator-agentdiet-dev-semantic-html-pruner-with-bm25-budget-8d7fc6e8 -d '<json body>'
```

Example prompt: Can you distill https://stripe.com/docs/api into clean Markdown, keeping it under 2000 tokens and focusing on the query 'authentication and API keys'?

## When to prefer this

Choose this endpoint when an AI agent needs to consume web page content efficiently without burning large token budgets on raw HTML boilerplate, navigation, and scripts. It is especially valuable when you need BM25-ranked relevance filtering (mode=budget with a query) to extract only the portions of a page relevant to a specific question, or when you want structured JSON tables or a skeleton outline rather than raw Markdown. Prefer it over generic scrapers when token cost savings and ROI proof matter, and when working within constrained LLM context windows.

## Known failure modes

- URL is unreachable or returns non-200 status — endpoint may return an error status
- HTML payload is empty or malformed — distillation may return minimal or empty data
- Token budget (max_tokens) set too low for meaningful content — excerpt may be truncated heavily
- BM25 query provided without mode=budget — query may be ignored
- Rate limiting or payment failure via x402 — request rejected before processing
- URL blocks headless fetching (bot protection) — content fetch fails

## How this service works

Distill a URL or raw HTML payload with BM25 budget control (Live x402 v2, IETF MPP & MCP endpoint at https://llms-generator.agentdiet.dev/v1/semantic-html-pruner)

## Output

Returns a JSON object with a 'data' field containing the distilled content (Markdown text, structured JSON tables/metadata, a skeleton outline, or a BM25-ranked budget excerpt depending on mode), plus an 'roi_proof' object showing raw_tokens, distilled_tokens, tokens_saved, compression_ratio_pct, llm_cost_saved_usd, and roi_multiplier, along with optional cache hit/miss status and billing details.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url"
 ],
 "properties": {
  "url": {
   "type": "string",
   "minLength": 1,
   "description": "Target webpage URL to fetch and distill (or builtin://stripe-docs)"
  },
  "html": {
   "type": "string",
   "description": "Optional raw HTML string to distill directly without fetching"
  },
  "mode": {
   "enum": [
    "markdown",
    "structured",
    "skeleton",
    "budget"
   ],
   "type": "string",
   "default": "markdown",
   "description": "Distillation mode"
  },
  "query": {
   "type": "string",
   "description": "BM25 focus query when mode=budget"
  },
  "max_tokens": {
   "type": "integer",
   "minimum": 20,
   "description": "Strict token budget ceiling when mode=budget"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "status",
  "roi_proof",
  "data"
 ],
 "properties": {
  "data": {
   "type": "object",
   "description": "Distilled Markdown, structured JSON tables/metadata, skeleton outline, or BM25 budget excerpt."
  },
  "cache": {
   "type": "object",
   "properties": {
    "status": {
     "enum": [
      "HIT",
      "MISS"
     ],
     "type": "string"
    },
    "cache_key": {
     "type": "string"
    },
    "latency_ms": {
     "type": "number"
    }
   }
  },
  "status": {
   "type": "string",
   "description": "Execution status ('ok')"
  },
  "billing": {
   "type": "object",
   "properties": {
    "agent_id": {
     "type": "string"
    },
    "payment_method": {
     "type": "string"
    },
    "fee_charged_usd": {
     "type": "number"
    }
   }
  },
  "roi_proof": {
   "type": "object",
   "properties": {
    "raw_tokens": {
     "type": "integer"
    },
    "tokens_saved": {
     "type": "integer"
    },
    "roi_multiplier": {
     "type": "number"
    },
    "distilled_tokens": {
     "type": "integer"
    },
    "llm_cost_saved_usd": {
     "type": "number"
    },
    "compression_ratio_pct": {
     "type": "number"
    }
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/llms-generator-agentdiet-dev-semantic-html-pruner-with-bm25-budget-8d7fc6e8/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from llms-generator.agentdiet.dev](https://www.zero.xyz/host/llms-generator.agentdiet.dev/llms.txt)