# WURK API Agent-to-Human Advanced Jobs (Base)

> WURK API Agent-to-Human Advanced Jobs (Base) is a paid API for AI agents from wurkapi.fun, paid per call via x402, $0.1/call, status unknown (last checked 2026-10-11).

Launches agent-to-human microtask jobs on Base network, hiring real humans for feedback, moderation, ranking, and other tasks paid in USDC via x402

## Facts

- Endpoint: GET https://wurkapi.fun/base/agenttohumanadvanced?utm_source=zero.xyz
- Price: $0.1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/wurk-api-agent-to-human-advanced-jobs-base-7849c5da
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_objQErMrWO0DqM_ZBzEyU

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 wurk-api-agent-to-human-advanced-jobs-base-7849c5da
```

Example prompt: I need to hire real humans to evaluate and rank three product descriptions I've written — create an advanced agent-to-human job on WURK with creator selection so I can pick the best responses, requiring xBlueVerified participants, and enable human verification.

## When to prefer this

Use this endpoint when an AI agent needs structured human judgment — feedback, ranking, moderation, or creative selection — with advanced controls like participant credentialing (X score, Seeker ecosystem, blue badge), proof-of-human gating, and manual winner selection. Prefer over quick jobs when task complexity warrants filtering for quality participants or when the agent needs to handpick winners from submissions.

## Known failure modes

- Invalid or missing action parameter returns error
- Insufficient USDC balance causes payment failure via x402
- Strict participant requirements (e.g. xBlueVerified + high tweetScoutScore) can cause very slow or incomplete job fill
- Complex prompts may take hours to days to receive sufficient submissions
- Invalid secret when calling choose-winners endpoint returns unauthorized error
- Malformed submissionIds array when selecting winners causes 400 error

## How this service works

WURK API helps AI agents hire real humans for feedback, moderation, ranking, and microtasks. Launch agent-to-human jobs on x402 and Tempo MPP using USDC on Solana and Base.

## Output

Returns a JSON object with ok/paid status, a unique jobId, a secret for job management, the network used (base), a statusUrl to poll for submissions, a policyGuide explaining requirement options and timing expectations, and instructions for choosing winners when selectionType=creator.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "required": {
   "type": "string"
  },
  "properties": {
   "type": "string",
   "description": {
    "type": "string",
    "description": "Required when action=create. Prompt/instructions for participants. Simpler prompts usually fill faster; complex or high-effort prompts can take much longer."
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "ok": true,
  "paid": true,
  "jobId": "abcd1234",
  "secret": "secret",
  "network": "base",
  "statusUrl": "https://wurkapi.fun/base/agenttohumanadvanced?action=view&secret=...",
  "policyGuide": "Use at most one requirement via the single \"requirement\" field (tweetScoutScore:0, tweetScoutScore:10, tweetScoutScore:25, xMetricScore:75, xMetricScore:250, seekerUser, xBlueVerified). Definitions: Sorsa=tweetScoutScore, WURK X profile score=xMetricScore, seekerUser=Solana Seeker ecosystem users, xBlueVerified=X blue-badge accounts. Requirement choices can increase minimum perUser and reduce maximum winners, and can slow submission volume. Optional human_verified=yes enables VeryAI proof-of-human gating (reverify_on=1 and human_verified_requirement=1), which narrows the participant pool and can slow fill speed. For creator-selection jobs, pick winners through /api/agenttohumanadvanced/choose-winners.",
  "chooseWinners": {
   "when": "Only when selectionType=creator",
   "method": "POST",
   "endpoint": "https://wurkapi.fun/api/agenttohumanadvanced/choose-winners",
   "requires": [
    "secret",
    "submissionIds"
   ],
   "submissionIds": "Array of submission ids or comma-separated string"
  },
  "selectionType": "creator",
  "human_verified": "yes",
  "timingExpectations": "Submission speed is variable: simple/broad jobs can receive many submissions in minutes, while complex tasks or strict requirements can take hours to days to fill.",
  "selectionTimeMinutes": 60
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/wurk-api-agent-to-human-advanced-jobs-base-7849c5da/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from wurkapi.fun](https://www.zero.xyz/host/wurkapi.fun/llms.txt)