← Back to Skills Marketplace
Anime Image Downloader (Safebooru, Download Only)
by
litousteven
· GitHub ↗
· v1.0.5
· MIT-0
316
Downloads
0
Stars
1
Active Installs
6
Versions
Install in OpenClaw
/install safebooru-fetcher
Description
Download multiple high-quality original anime images from Safebooru by tags, saving files locally with safe content only.
Usage Guidance
This skill appears consistent and low-risk for its stated downloader purpose, but take the usual precautions before running untrusted code: review the included Python files yourself, run inside a restricted environment or container if possible, and avoid running as root. Be aware it will write image files to ./downloads (so check disk use and the directory). The script relies on safebooru.org being SFW; if you need stricter rating controls, add explicit filters before downloading. Also note that an agent allowed to call skills autonomously could download many files if misused—limit counts or add rate limits when invoking.
Capability Analysis
Type: OpenClaw Skill
Name: safebooru-fetcher
Version: 1.0.5
The skill is a legitimate utility for searching and downloading images from Safebooru using its public DAPI. The code in `safebooru.py` and `safebooru_fetcher.py` is well-structured, implements proper filename sanitization via `_safe_slug`, and includes rate-limiting to respect the provider's service. No evidence of data exfiltration, unauthorized network calls, or prompt injection was found; the instructions in `SKILL.md` actually serve to restrict the agent's scope to local downloads only.
Capability Assessment
Purpose & Capability
Name/description claim a safebooru downloader and the code only accesses safebooru DAPI endpoints (POST_DAPI and TAG_DAPI). There are no unrelated network endpoints, binaries, or environment variables requested.
Instruction Scope
SKILL.md and the CLI/docs state the skill only downloads to a local ./downloads directory and does not upload or post elsewhere. The code implements that behavior. Minor note: SKILL.md says 'safe content only' (Safebooru is SFW), but the script does not add extra rating enforcement itself — it relies on the safebooru dataset and tag filters.
Install Mechanism
No install spec that pulls arbitrary code from the network; included files are Python source and a small requirements.txt (requests). Running the script writes files to disk as expected. No downloads from unknown hosts or archive extraction.
Credentials
The skill requests no environment variables, credentials, or config paths. The only external access is to safebooru.org per its stated purpose.
Persistence & Privilege
always is false and there are no persistent modifications to other skills or system configs. The skill can be invoked autonomously (platform default) and will write files to disk, which is expected for a downloader.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install safebooru-fetcher - After installation, invoke the skill by name or use
/safebooru-fetcher - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.5
- Internal metadata files (.clawhub/origin.json, _meta.json) have been removed.
- No user-facing functionality or documentation changes.
v1.0.4
- Updated documentation in SKILL.md to clarify usage and improve formatting
- No functional changes to the code or features
v1.0.3
- Added support for sorting, pagination, and minimum score filtering in image downloads
- Introduced tag auto-suggestion/completion via the --suggest option
- Updated documentation with new usage examples and expanded common tag lists
- Now only depends on Python and requests (dropped mandatory agent-browser dependency)
v1.0.2
- 增加了 agent-browser 作为必须依赖，并提供安装命令及说明原因
- 依赖列表和说明内容更清晰，便于快速配置使用
- 其它文档结构和功能说明无变化
v1.0.1
- Renamed the skill to "Anime Image Downloader" for broader clarity.
- Improved description to emphasize "图片搜索" (image search) capabilities.
- Enhanced feature list with clearer distinctions between supported and unsupported functions.
- Added a "wallpaper" tag example and a new "Keywords" section to improve searchability.
- Minor language and formatting improvements for clarity.
v1.0.0
Initial release of safebooru-fetcher.
- Fetches and downloads high-quality original images from Safebooru by tags.
- Supports multi-tag searches (space-separated) and batch downloads.
- Saves all images to a local directory (`./downloads/`).
- Designed for download only; does not send, upload, or post files.
- Dependence on Python 3, `requests`, and `agent-browser`.
- Outputs the number of images downloaded and the save path.
Metadata
Frequently Asked Questions
What is Anime Image Downloader (Safebooru, Download Only)?
Download multiple high-quality original anime images from Safebooru by tags, saving files locally with safe content only. It is an AI Agent Skill for Claude Code / OpenClaw, with 316 downloads so far.
How do I install Anime Image Downloader (Safebooru, Download Only)?
Run "/install safebooru-fetcher" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Anime Image Downloader (Safebooru, Download Only) free?
Yes, Anime Image Downloader (Safebooru, Download Only) is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Anime Image Downloader (Safebooru, Download Only) support?
Anime Image Downloader (Safebooru, Download Only) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Anime Image Downloader (Safebooru, Download Only)?
It is built and maintained by litousteven (@litousteven); the current version is v1.0.5.
More Skills