← Back to Skills Marketplace
yy756127197

Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0

by yy756127197 · GitHub ↗ · v2.0.0 · MIT-0
cross-platform ⚠ suspicious
158
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install yy756127197-doubao-image
Description
使用字节跳动豆包 Doubao SeeDream 模型生成高质量图片。支持文生图、AI 绘图、插画创作等功能。
Usage Guidance
Key things to consider before installing/running: 1) Metadata mismatch: the registry claims no required env vars, but the scripts require ARK_API_KEY — confirm the skill author/source and insist the package metadata be corrected to declare ARK_API_KEY as required. Don't rely solely on registry fields. 2) Inspect before running: you should review the shipped scripts locally (they are included) and confirm the API endpoint (ark.cn-beijing.volces.com) and model name match what you expect. The code is readable; if you can't inspect, avoid running. 3) Limit information exposure: the environment check prints current user, hostname, working dir, and the first 10 characters of ARK_API_KEY. That is convenient for debugging but leaks local info to whoever can see the terminal. Run the scripts in a trusted, isolated environment (or remove those printouts) if you are concerned. 4) Network fetch risk: the skill will download image URLs returned by the API — those URLs may point to external hosts and will be fetched without additional validation. If you must run it, consider running in a network-restricted environment or auditing the returned URLs before automatic downloading. 5) Secrets handling: supply ARK_API_KEY only to trusted code. Prefer setting the env var in a session (not checked into files) and verify there are no hard-coded keys (the package asserts none). 6) Source provenance: the skill lists 'Source: unknown' and placeholder GitHub links. Favor installing only from known/trusted sources or official repositories. Ask the publisher to provide a real homepage/repo and fix metadata inconsistencies. If you want, I can point out the exact lines in the scripts that print system info or the API key preview and suggest precise edits to harden them (e.g., remove whoami/pwd prints, stop displaying any part of the API key, add URL validation before downloading).
Capability Analysis
Type: OpenClaw Skill Name: yy756127197-doubao-image Version: 2.0.0 The `doubao-image` skill bundle is a well-structured and professional tool designed to generate AI images using the ByteDance Doubao SeeDream model via the official Volcengine ARK API. It includes both Bash and Python implementations (`doubao-image-generate.sh` and `doubao-image-generate.py`) that handle API authentication securely via environment variables and manage image downloads to a local directory. The code is transparent, follows security best practices (such as using Python for safe JSON handling in shell scripts), and lacks any signs of malicious intent, data exfiltration, or prompt injection. All files, including the extensive documentation and environment check scripts, are consistent with the stated purpose of the tool.
Capability Tags
cryptocan-make-purchases
Capability Assessment
Purpose & Capability
The name/description (Doubao SeeDream image generation) align with the provided code and runtime instructions: the Bash and Python scripts call a Volcengine/ARK API endpoint to generate and download images. However the registry metadata incorrectly lists no required environment variables and no install spec despite the package requiring an ARK_API_KEY and including runnable scripts — that metadata mismatch is an incoherence that should be resolved.
Instruction Scope
SKILL.md and the scripts confine behavior to: checking environment, calling POST https://ark.cn-beijing.volces.com/api/v3/images/generations with the ARK_API_KEY, parsing the JSON result, and downloading the returned image URL(s) to disk. The environment-check script prints system info (whoami, pwd, hostname) and displays the first 10 characters of ARK_API_KEY for verification; the scripts also perform ping/curl checks. They do not appear to access unrelated credentials or arbitrary local config paths. Two caution points: (1) printing partial API key and system details may leak sensitive local info to whoever runs the script output, and (2) downloaded image URLs come from API responses and may point to arbitrary hosts (the script will fetch them), so be mindful that network fetches could follow redirects or access external servers.
Install Mechanism
There is no remote installer or download URL; the repository provides local Bash and Python scripts and documentation. No installation step pulls arbitrary remote archives or executes unsigned code from external servers. Because the skill is 'instruction-only' in the registry but includes code files, the primary risk is running those local scripts — no high-risk installer was observed.
Credentials
The scripts legitimately require one secret: ARK_API_KEY (firewall/API key for Volcengine/ARK). That credential is proportional to the skill's purpose. However the skill metadata in the registry lists no required env vars or primary credential, which is inconsistent and could mislead users/automated reviewers. Additionally, the environment-check script prints the first 10 characters of the API key and exposes local system information (user, hostname, working directory) to stdout — this is not necessary for functionality and increases information exposure.
Persistence & Privilege
The skill does not request always:true or any platform-wide privileges. It writes generated images to a local output directory (configurable) and does not modify other skills or global agent settings. File writes are limited to the output directory. No evidence that the skill attempts to persist credentials or change system-wide config.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install yy756127197-doubao-image
  3. After installation, invoke the skill by name or use /yy756127197-doubao-image
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.0
v2.0.0: Refactored code, added Python implementation, improved error handling and documentation
Metadata
Slug yy756127197-doubao-image
Version 2.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0?

使用字节跳动豆包 Doubao SeeDream 模型生成高质量图片。支持文生图、AI 绘图、插画创作等功能。 It is an AI Agent Skill for Claude Code / OpenClaw, with 158 downloads so far.

How do I install Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0?

Run "/install yy756127197-doubao-image" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0 free?

Yes, Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0 support?

Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Doubao Image Generator — AI 文生图工具 based on SeeDream 5.0?

It is built and maintained by yy756127197 (@yy756127197); the current version is v2.0.0.

💬 Comments