# AgentToll LLM-Pack

> AgentToll LLM-Pack is a paid API for AI agents from agent-toll.agentfetcher.workers.dev, paid per call via x402, $0.00375/call, status unknown (last checked 2026-09-13).

Fetches a public URL and converts its content into token-budgeted, LLM-ready markdown suitable for model context or RAG pipelines

## Facts

- Endpoint: GET https://agent-toll.agentfetcher.workers.dev/v1/llm-pack
- Price: $0.00375/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agenttoll-llm-pack-8d1dd821
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_HGbiNfpi37rd-60a47--i

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 agenttoll-llm-pack-8d1dd821
```

Example prompt: Can you fetch https://docs.example.com/api-reference and convert it to clean LLM-ready markdown with a token budget of 8000 tokens so I can paste it into my RAG pipeline?

## When to prefer this

Prefer this endpoint over raw HTML fetching or generic scrapers when you need clean, token-budgeted markdown ready for direct injection into an LLM prompt or RAG system. Ideal when you want to avoid processing raw HTML yourself, need to control token consumption precisely, or are building an agentic pipeline where web content must be reliably compressible to a known size.

## Known failure modes

- URL is not publicly accessible or returns non-200 HTTP status
- URL points to non-text content (PDF, binary, etc.) that cannot be converted
- maxTokens value is outside the allowed range of 500–32000
- Payment not completed (x402 flow) after the first free request per IP
- Rate limiting or IP-based quota exhaustion
- Page content is too large or complex to process within timeout

## How this service works

x402 probe surface for llm-pack. Real usage: POST JSON {url, maxTokens?}. Unpaid GET/POST → 402 with PAYMENT-REQUIRED. Supports optional X-Latch-Bid-Usd.

## Output

Returns token-budgeted markdown derived from the fetched public URL, trimmed to fit within the specified maxTokens limit (500–32000, default 4000). The output is clean, model-ready text suitable for direct injection into LLM context windows or RAG systems.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Public URL to convert into token-budgeted LLM-ready markdown for model context / RAG"
  },
  "noCache": {
   "type": "boolean",
   "description": "Skip cache and re-fetch the page"
  },
  "maxTokens": {
   "type": "number",
   "description": "Token budget for packed markdown (500–32000, default 4000)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agenttoll-llm-pack-8d1dd821/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent-toll.agentfetcher.workers.dev](https://www.zero.xyz/host/agent-toll.agentfetcher.workers.dev/llms.txt)