# Agoranome Machine Buyer Panel

> Agoranome Machine Buyer Panel is a paid API for AI agents from api.agoranome.com, paid per call via x402, $1/call, status unknown (last checked 2026-10-28).

Simulates buyer agent decisions to test whether your x402 API listing beats rival offers, returning win-rate stats and actionable edit suggestions for description, price, and output examples.

## Facts

- Endpoint: POST https://api.agoranome.com/v1/seller/panel?utm_source=zero.xyz
- Price: $1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-28
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agoranome-machine-buyer-panel-bdd99f1e
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_OYC53smMY-7wNK2oRf9pH

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 agoranome-machine-buyer-panel-bdd99f1e -d '<json body>'
```

Example prompt: Run the Agoranome buyer panel simulation for my API at https://api.mything.com/v1/data with tasks ['find best price','compare descriptions'] and tell me the win rate and which fields to edit to beat the competition.

## When to prefer this

Choose this endpoint when you are an x402 API seller wanting a simulated, agent-perspective evaluation of your listing's competitiveness — specifically when you need a quantified win-rate with confidence intervals and field-level edit recommendations. It is more actionable than a static lint check because it models actual buyer agent decision-making rather than rule-based quality flags.

## Known failure modes

- Invalid or unreachable resource URL returns an error rejecting the listing
- tasks array with more than 3 items is rejected by schema validation
- Ambiguous or unrecognized task strings may produce inconclusive verdicts across all edits
- Payment failure or insufficient USDC balance blocks the call entirely
- Low sample size may yield wide confidence intervals making verdicts inconclusive

## How this service works

Will buyer agents choose my x402 API? The Machine Buyer Panel is an A/B conversion test: buyer-agent models choose between your offer and its real Bazaar rivals for up to three tasks, across buyer personas and list positions. Returns your win rate with confidence intervals, why you lost, and up to three truthful edits tested on the same trials.

## Output

Returns a JSON object with: a headline string summarizing how many of N simulated buyer decisions favored your listing; a baseline win_rate with point estimate and 95% confidence interval bounds; and an edits array where each item identifies the listing field tested (description, price, or output_example), the direction of change tested, and a verdict of better, worse, or inconclusive.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "tasks": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "maxItems": 3,
   "minItems": 1
  },
  "resource": {
   "type": "string",
   "description": "Your resource URL"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "result": {
   "edits": [
    {
     "kind": "<description|price|output_example>",
     "vs_baseline": {
      "verdict": "<better|worse|inconclusive>"
     }
    }
   ],
   "baseline": {
    "win_rate": {
     "ci95": [
      "<lo>",
      "<hi>"
     ],
     "rate": "<0-1>"
    }
   },
   "headline": "Buyer agents picked your offer in <wins> of <n> decisions."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agoranome-machine-buyer-panel-bdd99f1e/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.agoranome.com](https://www.zero.xyz/host/api.agoranome.com/llms.txt)