Agent Skills › opensquilla/opensquilla › nano-banana-pro-openrouter

nano-banana-pro-openrouter

GitHub

为Nano Banana Pro提供确定性的OpenRouter图像生成适配，用于meta-skill执行。支持JSON结构化槽位和纯文本模式，安全处理密钥并输出IMAGE_READY记录，避免非零退出错误。

src/opensquilla/skills/bundled/nano-banana-pro-openrouter/SKILL.md opensquilla/opensquilla

Trigger Scenarios

需要本地图像文件且不想启动LLM代理时通过stdin传入媒体槽位或提示词生成图像时

Install

npx skills add opensquilla/opensquilla --skill nano-banana-pro-openrouter -g -y

More Options

Non-standard path

npx skills add https://github.com/opensquilla/opensquilla/tree/main/src/opensquilla/skills/bundled/nano-banana-pro-openrouter -g -y

Use without installing

npx skills use opensquilla/opensquilla@nano-banana-pro-openrouter

指定 Agent (Claude Code)

npx skills add opensquilla/opensquilla --skill nano-banana-pro-openrouter -a claude-code -g -y

安装 repo 全部 skill

npx skills add opensquilla/opensquilla --all -g -y

预览 repo 内 skill

npx skills add opensquilla/opensquilla --list

SKILL.md

Frontmatter

{
    "name": "nano-banana-pro-openrouter",
    "homepage": "https:\/\/clawhub.ai\/skills\/nano-banana-pro-openrouter",
    "metadata": {
        "opensquilla": {
            "risk": "high",
            "requires": {
                "env": [],
                "bins": [
                    "python3"
                ],
                "config": [
                    "awesome_webpage.openrouter.api_key_env",
                    "awesome_webpage.openrouter.base_url",
                    "awesome_webpage.openrouter.models.image_generation",
                    "awesome_webpage.output_dir"
                ]
            },
            "capabilities": [
                "network-write",
                "filesystem-write"
            ]
        }
    },
    "entrypoint": {
        "env": {
            "{{ with.api_key_env | default('OPENROUTER_API_KEY') }}": "{{ with.api_key | default('') }}"
        },
        "args": [
            "--model",
            "{{ with.model | default('google\/gemini-3-pro-image-preview') }}",
            "--base-url",
            "{{ with.base_url | default('https:\/\/openrouter.ai\/api\/v1') }}",
            "--api-key-env",
            "{{ with.api_key_env | default('OPENROUTER_API_KEY') }}",
            "--output-dir",
            "{{ with.output_dir }}",
            "--filename",
            "{{ with.filename | default('image.png') }}",
            "--resolution",
            "{{ with.resolution | default('1K') }}",
            "--max-images",
            "{{ with.max_images | default('6') }}",
            "--local-path-prefix",
            "{{ with.local_path_prefix | default('project\/assets\/images') }}"
        ],
        "parse": "text",
        "stdin": "{{ with.payload | default(with.prompt | default(inputs.user_message)) }}",
        "command": "python {baseDir}\/scripts\/openrouter_image.py",
        "timeout": 300
    },
    "provenance": {
        "origin": "clawhub-mit0",
        "license": "MIT-0",
        "upstream_url": "https:\/\/clawhub.ai\/skills\/nano-banana-pro-openrouter",
        "maintained_by": "OpenSquilla"
    },
    "description": "Deterministic OpenRouter image generation adapter for Nano Banana Pro \/ Gemini image models. Use as skill_exec when a meta-skill needs local image files and structured IMAGE_READY records without spawning an LLM agent.",
    "user-invocable": false,
    "disable-model-invocation": true
}

Nano Banana Pro OpenRouter Adapter

This skill is a deterministic adapter around OpenRouter image generation. It is intended for meta-skill skill_exec use, not as an open-ended agent surface.

Contract

Uses an explicit with.api_key value by injecting it into the configured with.api_key_env child process environment variable; it never renders the key into argv.
Does not read .env files, prompt for credentials, print credentials, or write credentials to disk.
Uses only the model, base URL, output directory, and local path prefix passed by the caller.
Saves generated image bytes under the supplied output directory.
Emits one IMAGE_READY: JSON line per saved image.
On missing config or provider failure, emits IMAGE_CONFIG_NEEDED or IMAGE_GENERATION_FAILED instead of raising non-zero process errors.

Meta-Skill Payload Mode

When stdin is JSON containing media_slots, image_slots, slots, or page_outline, the adapter generates image slots and preserves already downloaded images. Structured slots are preferred over free-form outline text:

{
  "requirement_framing": "...",
  "media_slots": {"slots": [{"slot_id": "hero-visual", "modality": "image"}]},
  "page_outline": "...",
  "image_download": "...",
  "include_images": "YES",
  "visual_style": "..."
}

Existing IMAGE_READY: records in image_download are preserved. If IMAGE_DOWNLOAD_INCOMPLETE: lists unfilled_slot_ids, only those slots are generated. If the caller requested images but both structured slots and outline slot parsing are empty, the adapter synthesizes minimal webpage-safe image slots from the brief instead of emitting no_image_slots_to_generate.

Plain Prompt Mode

When stdin is plain text, the adapter generates one image using that text as the prompt and the --filename stem as the slot_id.

Version History

7f72a32 Current 2026-07-05 18:39

Metadata

Files: 0
Version: 7f72a32
Hash: e8b28c3f
Indexed: 2026-07-05 18:39

nano-banana-pro-openrouter