安全决策报告
image-generation-guide
A legitimate image generation guide with documented shell execution and file cleanup risks that are acknowledged in skill-card.md.
为什么得出这个结论
1/4 个维度触发 通过
声明与实际能力
声明资源与推断能力基本一致。
阻止
隐藏执行与外联
提取到 1 个高危 IOC 或外联信号。
通过
攻击链与高危发现
没有形成明确的恶意路径。
复核
依赖与供应链卫生
没有完整依赖信息，供应链判断需要保留弹性。
风险分是怎么被拉高的
curl|sh installation pattern +10
Present in guide but targets legitimate ollama.com; common but risky pattern
Hidden file cleanup behavior +12
flux-gen removes $HOME/*.png without clear prominence in SKILL.md (only in skill-card.md)
最关键的证据
中危 代码执行
Remote script execution via curl|sh
The full-guide.md includes `curl -fsSL https://ollama.com/install.sh | sh` which executes remote code. While the source is legitimate (ollama.com), this pattern is inherently dangerous and not declared in SKILL.md.
guide/full-guide.md:71 Prefer package managers (brew) or manual verification before execution. Document this in SKILL.md if intentional.
低危 文档欺骗
Undeclared file cleanup behavior
The flux-gen script removes PNG files from $HOME without warning. This hidden functionality is only documented in skill-card.md 'Known Risks' section, not in SKILL.md.
scripts/flux-gen:25 Either document this in SKILL.md or modify the script to only clean its own output directory.
提示 敏感访问
Accesses user home directory for output
The script creates and manages files in ~/Pictures/OpenClaw and searches /tmp and $HOME for generated images.
scripts/flux-gen:12 This is expected behavior for an image generation tool; not a security issue.
声明能力 vs 实际能力
命令执行 通过
声明 WRITE
→ 推断 WRITE
SKILL.md declares shell.exec; guide:71 shows curl|sh pattern 文件系统 通过
声明 NONE
→ 推断 WRITE
flux-gen:25-26 removes $HOME/*.png files 网络访问 通过
声明 READ
→ 推断 READ
SKILL.md declares http.request; curl downloads from ollama.com 可疑产物与外联
严重 危险命令
curl -fsSL https://ollama.com/install.sh | sh guide/full-guide.md:71
中危 外部 URL
https://thesolai.github.io/guides/signet-ai.html README.md:170
中危 外部 URL
https://thesolai.github.io/guides/openclaw-skills-guide.html README.md:171
中危 外部 URL
https://ollama.com/install.sh guide/full-guide.md:71
中危 外部 URL
https://discord.gg/ollama guide/full-guide.md:301
中危 外部 URL
https://ollama.com/blog/image-generation guide/full-guide.md:342
中危 外部 URL
https://blackforestlabs.ai/ guide/full-guide.md:343
中危 外部 URL
https://clawhub.ai/user/amrree skill-card.md:7
中危 外部 URL
https://clawhub.ai/amrree/skills/image-gen-guide skill-card.md:27
依赖与供应链
没有结构化依赖告警。
文件构成
7 个文件 · 741 行
Markdown 5 个文件 · 684 行Other 1 个文件 · 52 行JSON 1 个文件 · 5 行
需关注文件 · 4
guide/full-guide.md Remote script execution via curl|sh · curl -fsSL https://ollama.com/install.sh | sh · https://ollama.com/install.sh · https://discord.gg/ollama · https://ollama.com/blog/image-generation · https://blackforestlabs.ai/
README.md https://thesolai.github.io/guides/signet-ai.html · https://thesolai.github.io/guides/openclaw-skills-guide.html
skill-card.md https://clawhub.ai/user/amrree · https://clawhub.ai/amrree/skills/image-gen-guide
scripts/flux-gen Undeclared file cleanup behavior · Accesses user home directory for output
其他文件 · prompts.md · SKILL.md · _meta.json
安全亮点
No credential harvesting or API key theft observed
No data exfiltration or C2 communication detected
No base64 obfuscation or anti-analysis techniques
No access to sensitive paths like ~/.ssh, ~/.aws, or .env
No reverse shell or remote code execution backdoors
Known risks are documented in skill-card.md
Uses legitimate Ollama infrastructure (ollama.com)
No malicious dependencies or supply chain risks