# Generate Paid AI Images with BlockRun

AI assistants often need a direct way to create images during a coding or design session. This skill connects Claude, Codex, and Claude Code to BlockRun image generation using x402 USDC micropayments.

## Install

```bash
npx skillstore add blockrunai/nano-banana-blockrun
```

## Metadata

- - Slug: blockrunai-nano-banana-blockrun
- - Version: 1.0.0
- - Author: BlockRunAI
- - GitHub username: BlockRunAI
- - License: Apache-2.0
- - Repository: https://github.com/BlockRunAI/nano-banana-blockrun/tree/main/
- - Ref: main
- - Supported tools: Claude, Codex, Claude Code
- - Risk level: high
- - Risk factors: scripts, network, external\_commands, filesystem, env\_access
- - Quality score: 38
- - Quality tier: warning
- - Public page: https://skillstore.pages.dev/skills/blockrunai-nano-banana-blockrun
- - Manifest: https://skillstore.pages.dev/api/skills/blockrunai-nano-banana-blockrun/manifest

## Capabilities

- Generates images from text prompts through the BlockRun SDK.
- Supports google/nano-banana, google/nano-banana-pro, and openai/dall-e-3 model names.
- Uses x402 payment flow with USDC on the Base network.
- Reads a BLOCKRUN\_WALLET\_KEY value from environment configuration.
- Saves returned PNG data URLs as generated image files.
- Prints returned image URLs when the service does not return inline PNG data.

## Use Cases

- Create Concept Art: Generate quick visual concepts from prompts while exploring a product, game, or campaign idea.
- Produce App Mockup Assets: Create placeholder images for prototypes without leaving an AI coding session.
- Test Paid Image Workflows: Validate x402 payment-based image generation using a small wallet balance and explicit prompts.

## Prompt Templates

### Generate a Simple Image

```
Generate an image of a clean workspace with a laptop, notebook, and warm natural light.
```

### Choose a Specific Model

```
Generate an image of a futuristic city using google/nano-banana-pro and save it to my project assets folder.
```

### Create Product Visuals

```
Create three image prompt options for a modern fintech dashboard hero image, then generate the strongest option.
```

### Run a Controlled Paid Test

```
Before generating, confirm the selected model, output directory, and expected cost. Then create a 1024x1024 image of a Base network payment terminal.
```

## Limitations

- Requires USDC on Base and a wallet private key for payment signing.
- Depends on the external blockrun-llm Python package and BlockRun service availability.
- Generated files use predictable names and may overwrite previous outputs.
- This community skill has high security risk because it handles wallet credentials.

## Best Practices

- Use a dedicated wallet with only the small USDC balance needed for image generation.
- Review the model name, prompt, and expected cost before each paid request.
- Set an explicit output directory that does not contain unrelated .env files.

## Anti Patterns

- Do not use a wallet that holds production funds or unrelated assets.
- Do not store private keys in shared project folders or committed files.
- Do not run paid generations from untrusted prompts without reviewing the request.

## Security Audit

- - Safe to publish: false
- - Audited at: 2026-06-28T15:55:18.946\+00:00
- - Summary: Static critical heuristics are not confirmed as malicious, and no prompt injection attempt was found. However, the skill requires a wallet private key, loads .env files, signs paid x402 requests, and uses network calls through a third-party SDK. This is high risk for a community marketplace skill and should not be published without stronger review and user warnings.

## Stats

- - Views: 287
- - Downloads: 6
- - Favorites: 0
- - Popularity score: 0