agentleFS
0gfoundation0gfoundation/0g-agent-skills/skills/compute/text-to-image/SKILL.md

text-to-image

Skill15 starsChanged 58 days agoMITView on GitHub

Generate images from text prompts using 0G Compute Network providers (serviceType: text-to-image). Supports multiple resolutions and batch generation. To edit an existing image rather than generate a new one, use Image Editing (serviceType: image-editing) instead. Discover at runtime; do not hardcode. At last verification 0G mainnet had one text-to-image provider whose default model was z-image-turbo. Testnet had no text-to-image provider — check availability before targeting testnet. Image models are billed per generated image rather than per token. Look at m.pricing?.image…

  • Reads credentials
  • Sends data out

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call processResponse() after every generation

    line 49
  2. Very specific:

    Use correct processResponse() param order: (providerAddress, chatID, usageData)

    line 50
  3. Distinctive:

    Start with smaller dimensions for cost-effective testing

    line 52
  4. Distinctive:

    Use ethers v5 syntax

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Text-to-Image Generation

Metadata

  • Category: compute
  • SDK: @0gfoundation/0g-compute-ts-sdk ^0.9.0, ethers 6.13.1
  • Activation Triggers: "generate image", "text-to-image", "image generation", "create image"

Purpose

Generate images from text prompts using 0G Compute Network providers (serviceType: text-to-image). Supports multiple resolutions and batch generation.

To edit an existing image rather than generate a new one, use Image Editing (serviceType: image-editing) instead.

Models

Discover at runtime; do not hardcode. At last verification 0G mainnet had one text-to-image provider whose default model was z-image-turbo. Testnet had no text-to-image provider — check availability before targeting testnet.

Image models are billed per generated image rather than per token. Look at m.pricing?.image / m.pricing_usd?.image from getProviderModels().

Prerequisites

  • Node.js >= 20
  • @0gfoundation/0g-compute-ts-sdk and ethers installed
  • Funded and acknowledged provider with text-to-image service
  • .env with PRIVATE_KEY, RPC_URL, PROVIDER_ADDRESS

Quick Workflow

  1. Initialize broker
  2. Get service metadata (endpoint, model)
  3. Build request body with prompt, size, and count
  4. Generate auth headers with request body (required for signing)
  5. Make image generation request
  6. Extract ChatID from ZG-Res-Key header ONLY (no body fallback)
  7. Call processResponse(providerAddress, chatID) — no usage data needed

Core Rules

ALWAYS

  • Include request body when generating auth headers (signing requirement)
  • Get ChatID from ZG-Res-Key header only (no body fallback for images)
  • Call processResponse() after every generation
  • Use correct processResponse() param order: (providerAddress, chatID, usageData)
  • Acknowledge provider before first use
  • Start with smaller dimensions for cost-effective testing

NEVER

  • Skip including the body in getRequestHeaders() for image requests
  • Try to extract ChatID from the response body (images only have header)
  • Generate large batches without checking balance first
  • Hardcode private keys
  • Use ethers v5 syntax

Code Examples

Basic Image Generation

import { ethers } from 'ethers';
import { createZGComputeNetworkBroker } from '@0gfoundation/0g-compute-ts-sdk';
import * as fs from 'fs';
import 'dotenv/config';

async function generateImage(prompt: string, size = '1024x1024'): Promise<string> {
  const provider = new ethers.JsonRpcProvider(process.env.RPC_URL);
  const wallet = new ethers.Wallet(process.env.PRIVATE_KEY!, provider);
  const broker = await createZGComputeNetworkBroker(wallet);

  const providerAddress = process.env.PROVIDER_ADDRESS!;
  const { endpoint, model } = await broker.inference.getServiceMetadata(providerAddress);

  const requestBody = { model, prompt, n: 1, size };

  // IMPORTANT: Include body for request signing
  const headers = await broker.inference.getRequestHeaders(
    providerAddress,
    JSON.stringify(requestBody),
  );

  const response = await fetch(`${endpoint}/images/generations`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json', ...headers },
    body: JSON.stringify(requestBody),
  });

  const data = await response.json();
  const imageUrl = data.data[0].url;

  // ChatID from header ONLY for images
  const chatID = response.headers.get('ZG-Res-Key') || response.headers.get('zg-res-key');

  if (chatID) {
    await broker.inference.processResponse(providerAddress, chatID);
  }

  return imageUrl;
}

// Usage
const imageUrl = await generateImage('A futuristic city skyline at sunset');
console.log('Image URL:', imageUrl);

Generate and Save Locally

async function generateAndSave(
  prompt: string,
  outputPath: string,
  size = '512x512',
): Promise<void> {
  const provider = new ethers.JsonRpcProvider(process.env.RPC_URL);
  const wallet = new ethers.Wallet(process.env.PRIVATE_KEY!, provider);
  const broker = await createZGComputeNetworkBroker(wallet);

  const providerAddress = process.env.PROVIDER_ADDRESS!;
  const { endpoint, model } = await broker.inference.getServiceMetadata(providerAddress);

  const requestBody = { model, prompt, n: 1, size };
  const headers = await broker.inference.getRequestHeaders(
    providerAddress,
    JSON.stringify(requestBody),
  );

  const response = await fetch(`${endpoint}/images/generations`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json', ...headers },
    body: JSON.stringify(requestBody),
  });

  const data = await response.json();

  // Process response for fee settlement
  const chatID = response.headers.get('ZG-Res-Key') || response.headers.get('zg-res-key');
  if (chatID) {
    await broker.inference.processResponse(providerAddress, chatID);
  }

  // Download and save image
  const imageData = data.data[0];
  if (imageData.b64_json) {
    fs.writeFileSync(outputPath, Buffer.from(imageData.b64_json, 'base64'));
  } else if (imageData.url) {
    const imgResponse = await fetch(imageData.url);
    const buffer = Buffer.from(await imgResponse.arrayBuffer());
    fs.writeFileSync(outputPath, buffer);
  }

  console.log(`Image saved to ${outputPath}`);
}

Batch Generation

async function batchGenerate(
  prompts: string[],
  outputDir: string,
  size = '512x512',
): Promise<string[]> {
  const provider = new ethers.JsonRpcProvider(process.env.RPC_URL);
  const wallet = new ethers.Wallet(process.env.PRIVATE_KEY!, provider);
  const broker = await createZGComputeNetworkBroker(wallet);

  const providerAddress = process.env.PROVIDER_ADDRESS!;
  const { endpoint, model } = await broker.inference.getServiceMetadata(providerAddress);

  if (!fs.existsSync(outputDir)) fs.mkdirSync(outputDir, { recursive: true });

  const results: string[] = [];

  for (let i = 0; i < prompts.length; i++) {
    console.log(`Generating ${i + 1}/${prompts.length}: "${prompts[i]}"`);

    const requestBody = { model, prompt: prompts[i], n: 1, size };
    const headers = await broker.inference.getRequestHeaders(
      providerAddress,
      JSON.stringify(requestBody),
    );

    const response = await fetch(`${endpoint}/images/generations`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json', ...headers },
      body: JSON.stringify(requestBody),
    });

    const data = await response.json();
    const chatID = response.headers.get('ZG-Res-Key') || response.headers.get('zg-res-key');
    if (chatID) {
      await broker.inference.processResponse(providerAddress, chatID);
    }

    const outputPath = `${outputDir}/image-${i + 1}.png`;
    if (data.data[0].url) {
      const imgResponse = await fetch(data.data[0].url);
      fs.writeFileSync(outputPath, Buffer.from(await imgResponse.arrayBuffer()));
    }

    results.push(outputPath);
  }

  return results;
}

Error Handling

async function safeGenerateImage(prompt: string, size = '512x512'): Promise<string | null> {
  const provider = new ethers.JsonRpcProvider(process.env.RPC_URL);
  const wallet = new ethers.Wallet(process.env.PRIVATE_KEY!, provider);
  const broker = await createZGComputeNetworkBroker(wallet);

  const providerAddress = process.env.PROVIDER_ADDRESS!;

  try {
    const { endpoint, model } = await broker.inference.getServiceMetadata(providerAddress);
    const requestBody = { model, prompt, n: 1, size };

    const headers = await broker.inference.getRequestHeaders(
      providerAddress,
      JSON.stringify(requestBody),
    );

    const response = await fetch(`${endpoint}/images/generations`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json', ...headers },
      body: JSON.stringify(requestBody),
    });

    if (!response.ok) {
      throw new Error(`HTTP ${response.status}: ${await response.text()}`);
    }

    const data = await response.json();
    const chatID = response.headers.get('ZG-Res-Key') || response.headers.get('zg-res-key');

    if (chatID) {
      await broker.inference.processResponse(providerAddress, chatID);
    }

    return data.data[0].url || null;
  } catch (error) {
    console.error('Image generation failed:', error);
    return null;
  }
}

Supported Sizes

SizeCost (approx.)Best For
256x256~0.001 0GThumbnails, testing
512x512~0.002 0GStandard images
1024x1024~0.003 0GHigh quality

Anti-Patterns

// BAD: Missing body in getRequestHeaders for images
const headers = await broker.inference.getRequestHeaders(providerAddress);
// Must include: getRequestHeaders(providerAddress, JSON.stringify(requestBody))

// BAD: Trying to get chatID from response body
const chatID = data.id; // WRONG for images — use header only

// BAD: Large batch without balance check
await batchGenerate(hundredPrompts, './out'); // May run out of funds

// BAD: Hardcoding private keys
const wallet = new ethers.Wallet('0xabc123...', provider); // NEVER do this

// BAD: ethers v5 syntax
const provider = new ethers.providers.JsonRpcProvider(url); // v5!

Common Errors & Fixes

ErrorCauseFix
Insufficient balanceSub-account emptyTransfer more funds
Invalid request headersMissing body in header generationInclude body in getRequestHeaders()
Fee verification failedMissing chatIDEnsure ZG-Res-Key header exists
invalid sizeUnsupported dimensionsUse 256x256, 512x512, or 1024x1024

Related Skills

  • Provider Discovery — find image providers
  • Account Management — fund accounts
  • Compute + Storage — generate and store

References

  • Compute Patterns
  • Network Config

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.