# AgentDiet LLM Guard — Context Window Distiller & Token Diet

> AgentDiet LLM Guard — Context Window Distiller & Token Diet is a paid API for AI agents from status.agentdiet.dev, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-22).

Scans and compresses LLM context/text input to strip bloat, achieving up to 85-91% token reduction while preserving semantic content

## Facts

- Endpoint: GET https://status.agentdiet.dev/v1/llm-guard?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-22
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agentdiet-llm-guard-context-window-distiller-token-diet-03a6158b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_h-Iq7f91rzD6pafhXcX2f

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 agentdiet-llm-guard-context-window-distiller-token-diet-03a6158b
```

Example prompt: Run this context through AgentDiet's LLM Guard to strip it down — it's getting too long and expensive to send to the model as-is: 'You are a helpful assistant. The user has asked about the history of the Roman Empire...' Give me the compressed version and tell me the compression ratio.

## When to prefer this

Choose this endpoint when you need deterministic, measurable token compression with a verifiable ROI proof (compression_ratio_pct) before sending context to an LLM. Ideal for cost-sensitive agentic pipelines on x402/USDC pay-per-call infrastructure where you want to minimize inference spend per call. Prefer over generic summarization when you need a compression ratio metric rather than a human-readable summary.

## Known failure modes

- Missing required 'input' field returns validation error
- Invalid HTTP method enum value rejected
- Text too short to compress meaningfully may return low or zero compression ratio
- Payment failure via x402/USDC blocks execution
- Empty or null text query param may return error or no-op result

## How this service works

Deterministic B2A/A2A/MCP/x402/MPP/L402 execution and intelligence node at https://status.agentdiet.dev settling in USDC on Base.

## Output

Returns a JSON object with status 'ok', a data field containing the distilled/compressed output, and an roi_proof object showing the compression_ratio_pct (e.g. 91.4%), confirming how much token reduction was achieved.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET",
      "POST"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object",
     "properties": {
      "text": {
       "type": "string"
      }
     }
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string"
    },
    "example": {
     "type": "object"
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "data": {},
  "status": "ok",
  "roi_proof": {
   "compression_ratio_pct": 91.4
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agentdiet-llm-guard-context-window-distiller-token-diet-03a6158b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from status.agentdiet.dev](https://www.zero.xyz/host/status.agentdiet.dev/llms.txt)