扫描报告
5 /100
ai-commentary-zh
Sparki AI video editing skill with commentary/narrative focus, Chinese localization
This is a legitimate video editing CLI skill for Sparki AI that strictly adheres to its declared permissions with no malicious behavior.
可以安装
This skill is safe to use. No action required.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | READ | READ | ✓ 一致 | _meta.json: fs.read=[$CWD]; cli.py: Path.exists(), f.stat().st_size |
| 文件系统 | WRITE | WRITE | ✓ 一致 | _meta.json: fs.write=[$HOME/.openclaw/config, $HOME/.openclaw/workspace/sparki/v… |
| 网络访问 | READ | READ | ✓ 一致 | _meta.json: network.domains=[agent-api.sparki.io]; client.py: all httpx calls go… |
| 命令执行 | NONE | NONE | — | cli.py: No subprocess, os.system, or shell=True usage |
| 环境变量 | NONE | READ | ✓ 一致 | config.py: Reads SPARKI_API_KEY env var for legitimate authentication; no enumer… |
| 技能调用 | NONE | NONE | — | No inter-skill invocation observed |
7 项发现
中危 外部 URL 外部 URL
https://img.shields.io/badge/ClawHub-Skill-blueviolet README.md:3 中危 外部 URL 外部 URL
https://clawhub.io README.md:3 中危 外部 URL 外部 URL
https://img.shields.io/badge/version-1.0.12-blue README.md:4 中危 外部 URL 外部 URL
https://sparki.io SKILL.md:17 中危 外部 URL 外部 URL
https://agent-api.sparki.io src/sparki_cli/constants.py:61 中危 外部 URL 外部 URL
https://t.me/Sparki_AI_bot/upload src/sparki_cli/constants.py:62 中危 外部 URL 外部 URL
https://sparki.io/pricing src/sparki_cli/constants.py:101 目录结构
11 文件 · 33.2 KB · 1009 行 Python 7f · 904L
Markdown 2f · 70L
TOML 1f · 31L
JSON 1f · 4L
├─
▾
src
│ └─
▾
sparki_cli
│ ├─
__init__.py
Python
│ ├─
cli.py
Python
│ ├─
client.py
Python
│ ├─
config.py
Python
│ ├─
constants.py
Python
│ ├─
models.py
Python
│ └─
output.py
Python
├─
_meta.json
JSON
├─
pyproject.toml
TOML
├─
README.md
Markdown
└─
SKILL.md
Markdown
依赖分析 3 项
| 包名 | 版本 | 来源 | 已知漏洞 | 备注 |
|---|---|---|---|---|
typer | >=0.9.0 | pip | 否 | Minimum version pinned |
httpx | >=0.27.0 | pip | 否 | Minimum version pinned |
pydantic | >=2.0.0 | pip | 否 | Minimum version pinned |
安全亮点
✓ No shell execution (subprocess, os.system, shell=True) — all operations use Python standard libraries
✓ No obfuscation (base64, atob, eval) — code is human-readable
✓ No credential theft — API key used only for authenticated API calls to sparki.io
✓ No data exfiltration — all network traffic goes to declared domain agent-api.sparki.io
✓ No sensitive path access (~/.ssh, ~/.aws, .env) — only accesses declared config/output directories
✓ No supply chain risk — dependencies are well-known: typer, httpx, pydantic with pinned minimum versions
✓ No hidden functionality — SKILL.md accurately describes behavior; all features traceable to source code
✓ API key check priority: environment variable > config file, a standard secure pattern
✓ Config directory creation uses parents=True, exist_ok=True to avoid race conditions
✓ All HTTP errors caught and surfaced as NETWORK_ERROR, no silent failures