Relay Image Gen

What it does

When your agent uses it

• : creating images
• Generating pictures/illustrations

Example prompts

• “generate image”
• “create image”
• “gpt-image”
• “/relay-image-gen”

Requirements

• Python 3
• A Bash shell
• A credential in VECTORNODE_API_KEY
• A credential in QINGYUN_API_KEY

Workflow steps

5 steps, taken from the step headings in SKILL.md.

1. 1vectronode（首选 — 2026-06-17 起）
2. 2qingyun Gemini
3. 3OpenClaw image_generate（官方）
4. 4编辑 PROVIDER_CONFIG
5. 5添加到优先级链

What it can do on your machine

Read from SKILL.md and the folder at commit 7cefd81. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 2 files in scripts/ (Shell and Python), which the agent can run.

  Shell commands in SKILL.md call:

  • uv
  • python3
  • bash

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • VECTORNODE_API_KEY
  • XINGJIABIAPI_KEY
  • QINGYUN_API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Relay Image Gen loads about 1.4k tokens when it runs, and up to ~2k if it reads all its reference files. Until then it costs about 137 tokens; SKILL.md has 215 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~137

When it runs · the whole SKILL.md, loaded when a task matches

~1.4k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NoteMentions a .env fileSKILL.md:53

  要求**：`VECTORNODE_API_KEY` 在 `~/.openclaw/.env`

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from aAAaqwq/AGI-Super-Team at commit 7cefd81, republished under its MIT licence (© aAAaqwq). 215 words, ~1,436 tokens.

name

relay-image-gen

description

Multi-provider image generation with automatic priority fallback Multi-provider image generation with priority fallback. Priority: vectronode (gpt-image-2/gpt-image-2-all) → qingyun Gemini → OpenClaw image tool → relay providers. Supports 1K/2K/4K, multiple aspect ratios, auto fallback. Extensible: add custom relay/API providers to PROVIDER_CONFIG. Use when: creating images, generating pictures/illustrations, poster art. Trigger: "生成图片", "生成封面", "画一张", "generate image", "create image", "gpt-image", "gpt生图". Author: Daniel Li

1. vectronode (gpt-image-2 / gpt-image-2-all) ← 首选，VectorNode中转站，API Key已配，最可靠
   ↓ 失败/不可用
2. qingyun Gemini (gemini-3-pro-image-preview) ← 质量最佳
   ↓ 失败/不可用
3. OpenClaw image_generate tool ← 官方内置
   ↓ 失败/不可用
4. boluobao (中转API) ← relay fallback 1
   ↓ 失败/不可用
5. xingjiabi (多模型fallback) ← relay fallback 2
   ↓ 失败/不可用
6. 自定义扩展 providers ← 按需添加

python3 ~/.openclaw/skills/vectronode-image/scripts/vectronode_image.py generate \
  --prompt "your prompt" \
  --output "/path/to/output.png" \
  --size 2048x1152

export QINGYUN_API_KEY=$(pass show api/qingyun | head -n 1)
bash ~/clawd/skills/qingyun-api/scripts/qingyun-image-gemini.sh \
  "your prompt" \
  --ratio 3:4 \
  -o "/path/to/output"

image_generate(
  prompt="your prompt",
  size="1024x1024",
  aspectRatio="1:1"
)

uv run ~/.openclaw/skills/relay-image-gen/scripts/relay_image_gen.py \
  -p "your prompt" \
  -f "/path/to/output.jpg" \
  -a "3:4" \
  -r "1k"

~/clawd/projects/MediaClaw/output/articles/{YYYY-MM-DD}/{topic-slug}.{ext}

PROVIDER_CONFIG["my-provider"] = {
    "base_url": "https://api.my-provider.com/v1",
    "default_model": "my-model-name",
    "fallback_models": ["fallback-model-1", "fallback-model-2"],
    "env_key": "MY_PROVIDER_API_KEY",       # 环境变量名
    "pass_path": "api/my-provider",          # 或 pass 路径（二选一）
}

RELAY_PRIORITY = [
    {"name": "boluobao", "enabled": True},
    {"name": "my-provider", "enabled": True},  # ← 插入
    {"name": "xingjiabi", "enabled": True},
]

# 首选：vectronode GPT Image（当前最可靠）
python3 ~/.openclaw/skills/vectronode-image/scripts/vectronode_image.py generate \
  --prompt "Editorial flat illustration of AI tools comparison..." \
  --output ~/clawd/output/cover.png \
  --size 2048x1152

# 内容封面（备选：qingyun Gemini）
export QINGYUN_API_KEY=$(pass show api/qingyun | head -n 1)
bash ~/clawd/skills/qingyun-api/scripts/qingyun-image-gemini.sh \
  "Editorial flat illustration of AI tools comparison..." \
  --ratio 3:4 \
  -o "~/clawd/projects/MediaClaw/output/articles/2026-04-15/ai-tools-cover.png"

# 通用图片（relay fallback）
uv run ~/.openclaw/skills/relay-image-gen/scripts/relay_image_gen.py \
  -p "A serene Japanese garden with cherry blossoms" \
  -f "~/clawd/projects/MediaClaw/output/articles/2026-04-15/garden-bg.jpg" \
  -a "16:9" -r "2k"

# 强制指定 provider
uv run ~/.openclaw/skills/relay-image-gen/scripts/relay_image_gen.py \
  -p "Robot waving hello" \
  -f "robot.jpg" \
  -P xingjiabi -m gpt-image-1

# OpenClaw 内置（无bash）
image_generate(prompt="sunset over mountains", size="1792x1024", aspectRatio="16:9")

cat ~/clawd/skills/content-cover-gen/SKILL.md

© aAAaqwq, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (scripts, references) in skills/relay-image-gen of aAAaqwq/AGI-Super-Team.

• SKILL.md
• references/providers.md
• scripts/img-gen.sh
• scripts/relay_image_gen.py

Open the folder on GitHubat commit 7cefd81