← Back to Skills Marketplace
170
Downloads
0
Stars
1
Active Installs
8
Versions
Install in OpenClaw
/install xqx-image-generator
Description
使用火山引擎Ark进行文生图，支持生成头像、封面、海报、配图等，适用于用户请求包含‘生成图片’、‘AI绘图’等关键词的场景
Usage Guidance
This skill appears to do what it says: it sends user prompts to the Ark images API and returns an image URL. Before installing, consider: (1) you must provide ARK_API_KEY → this key will be sent to Ark's API (do not reuse a high-privilege key or share it in public docs); (2) long prompts are written to a fixed file in the agent's workspace (wdatas/xqx-img-prompt.txt) and the script prints the prompt to stdout — ensure the workspace and logs are private if prompts contain sensitive data; (3) the skill returns signed/temporary image URLs only — downloading/storing images is left to other workflows; (4) prefer injecting the API key via a secure secret manager or CI secret, and rotate the key if compromised. Overall the package is internally consistent and matches its stated purpose.
Capability Analysis
Type: OpenClaw Skill
Name: xqx-image-generator
Version: 1.0.7
The skill is a legitimate tool for generating images via the Volcengine Ark (Doubao) API. It uses the Python standard library to interact with the official API endpoint (ark.cn-beijing.volces.com) and correctly handles sensitive credentials via environment variables. The instructions for the agent to use a temporary file for long prompts are a practical design choice to avoid shell command length limits and potential injection risks, and no malicious behaviors or exfiltration patterns were identified.
Capability Assessment
Purpose & Capability
Name/description, declared env vars (ARK_API_KEY, ARK_IMAGE_MODEL), and included script all align: the skill posts prompt+model to Ark's images API and returns a URL. No unrelated credentials, binaries, or capabilities are requested.
Instruction Scope
SKILL.md requires the agent to write the user's prompt to a fixed workspace path (wdatas/xqx-img-prompt.txt) and run the script with that workspace as cwd. This is consistent with supporting long prompts, but it means prompts are written to disk in the agent's workspace and the script prints the prompt to stdout before making the API call (so prompt content appears in logs). The instructions do not read or transmit unrelated files or secrets.
Install Mechanism
No install spec or external downloads; included Python script uses only the standard library and an empty requirements.txt. Nothing is written to disk beyond the script itself and the prompt file that the agent is explicitly instructed to create.
Credentials
Only ARK_API_KEY and ARK_IMAGE_MODEL are required, which are appropriate for calling Ark's API. The script reads only those env vars. Note: ARK_API_KEY is sensitive and will be sent as a Bearer token to Ark's endpoint as expected.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges or modify other skills' configs. It does require agent invocation to write/read workspace files, which is normal for this use case.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install xqx-image-generator - After installation, invoke the skill by name or use
/xqx-image-generator - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.7
- 更新描述，使其更贴合用户请求“生成图片”“AI绘图”类场景，明确支持头像、封面、海报、配图等各类图片生成需求
- 其他内容未更动（无功能和用法更改）
v1.0.6
- Clarified documentation to state that the skill only returns accessible image URLs and does not handle image downloading or file saving.
- Updated description for broader application scenarios (e.g., profile pictures, covers, posters, illustrations).
- Improved output section to specify that results are always URLs and that saving images is the responsibility of the caller or workflow.
- No changes to code or file structure.
v1.0.5
**Major update: Adds explicit OpenClaw agent integration and --file mode support.**
- Introduced detailed instructions for OpenClaw agent workflows, including file-based prompt handling via `wdatas/xqx-img-prompt.txt`.
- Added mandatory `ARK_IMAGE_MODEL` environment variable; removed fallback/default model logic.
- Added `--file` (or `-f`) mode to support long, structured or multi-line prompts by reading from a predefined file.
- Updated documentation and usage examples for both command-line and agent/automated scenarios.
- Clarified output usage—respond directly with generated image URL.
- Minor refinements to skill description and metadata, emphasizing OpenClaw and agent compatibility.
v1.0.4
xqx-image-generator 1.0.4
- Updated description wording for clarity.
- Clarified default value for ARK_IMAGE_MODEL as doubao-seedream-5-0-260128.
- Marked ARK_IMAGE_MODEL environment variable as optional.
- Minor phrasing adjustments for environment setup guidance.
v1.0.3
- Added a dedicated metadata block specifying required environment variables (`ARK_API_KEY`, `ARK_IMAGE_MODEL`), with `ARK_API_KEY` as primary.
- Improved documentation on how to securely provide environment variables, emphasizing best practices and security warnings.
- Updated the environment variable instructions for clarity and safety.
- No changes to the code or functionality; documentation improvements only.
v1.0.2
- Removed the file .git_commit_msg.txt from the project.
- No functional changes to the skill itself.
v1.0.1
- Removed Feishu message sending feature and related parameters.
- Cleaned up documentation: Feishu options and user targeting removed from usage, examples, and output.
- Updated environment variable section for clarity.
- Minor file structure changes: removed _meta.json, added .git_commit_msg.txt.
v1.0.0
xqx-image-generator 1.0.0 – Initial Release
- Generate images from text using Ark (火山引擎豆包) via an OpenAI-compatible API.
- Optionally send generated images directly to Feishu users.
- Flexible size options including presets like 2K, 4K, or custom resolutions.
- Reads configuration from environment variables or a workspace TOOLS.md file.
- Requires only Python standard library, no third-party dependencies.
Metadata
Frequently Asked Questions
What is Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows.?
使用火山引擎Ark进行文生图，支持生成头像、封面、海报、配图等，适用于用户请求包含‘生成图片’、‘AI绘图’等关键词的场景. It is an AI Agent Skill for Claude Code / OpenClaw, with 170 downloads so far.
How do I install Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows.?
Run "/install xqx-image-generator" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows. free?
Yes, Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows. is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows. support?
Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows. is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Use Ark (火山引擎豆包) for text-to-image (文生图) via an OpenAI-compatible API. Suitable for short captions and long, structured creative or poster-style briefs in agent workflows.?
It is built and maintained by hidebug (@hidebug); the current version is v1.0.7.
More Skills