This report was generated in Chinese. Some content may be in Chinese.
Low Risk — Risk Score 20/100
Last scan:4 hr ago Rescan
20 /100
improvement-discriminator
对 AI Agent 技能改进候选进行多维度评分和排序,支持 LLM-as-Judge 和多审阅者盲审
合法的 AI 技能评分引擎,存在硬编码演示密码但不影响生产安全
Skill Nameimprovement-discriminator
Duration59.4s
Enginepi
ClawHub improvement-discriminator v1.1.0 by lanyasheng
📥 26
ClawHub Verdict Suspicious dynamic_code_executionllm_suspiciousvt_suspicious
Safe to install
清理测试文件中的硬编码凭证,使用环境变量替代演示密码

Findings 2 items

Severity Finding Location
Low
测试文件包含硬编码演示密码 Supply Chain
tests/test_p2a_integration.py:79 使用明文 'demo_password_123' 作为测试数据的加密密钥
key = hashlib.sha256("demo_password_123".encode()).digest()
→ 将测试密码移至环境变量或使用 pytest fixture 管理
tests/test_p2a_integration.py:68
Low
演示密码命名暗示生产用途 Doc Mismatch
interfaces/critic_engine.py:685 的 'DEMO_ONLY_NOT_FOR_PRODUCTION' 命名清晰表明为演示用,但与安全关键字匹配
password="DEMO_ONLY_NOT_FOR_PRODUCTION"
→ 虽然命名正确标识为演示,但建议使用环境变量统一管理
interfaces/critic_engine.py:685
ResourceDeclaredInferredStatusEvidence
Filesystem NONE READ ✓ Aligned scripts/score.py: 从 JSON 文件读取候选数据
Network NONE READ ✓ Aligned interfaces/llm_judge.py: 通过 SDK 调用 Claude/OpenAI API
Shell NONE NONE 无 shell 执行代码
2 High 2 findings
🔑
High API Key 疑似硬编码凭证
password="DEMO_ONLY_NOT_FOR_PRODUCTION"
interfaces/critic_engine.py:685
🔑
High API Key 疑似硬编码凭证
password="demo_password_123"
tests/test_p2a_integration.py:79

File Tree

14 files · 203.1 KB · 5900 lines
Python 12f · 5816L Markdown 2f · 84L
├─ 📁 interfaces
│ ├─ 🐍 __init__.py Python 49L · 1.1 KB
│ ├─ 🐍 assertions.py Python 569L · 17.4 KB
│ ├─ 🐍 critic_engine.py Python 1144L · 40.0 KB
│ ├─ 🐍 external_regression.py Python 672L · 21.2 KB
│ ├─ 🐍 human_review.py Python 638L · 20.6 KB
│ └─ 🐍 llm_judge.py Python 240L · 9.2 KB
├─ 📁 scripts
│ ├─ 🐍 rubric_evidence.py Python 289L · 11.2 KB
│ └─ 🐍 score.py Python 675L · 25.7 KB
├─ 📁 tests
│ ├─ 🐍 test_llm_judge.py Python 417L · 15.3 KB
│ ├─ 🐍 test_p1_integration.py Python 344L · 10.6 KB
│ ├─ 🐍 test_p2a_integration.py Python 465L · 15.5 KB
│ └─ 🐍 test_score.py Python 314L · 12.0 KB
├─ 📝 README.md Markdown 3L · 88 B
└─ 📝 SKILL.md Markdown 81L · 3.1 KB

Dependencies 2 items

PackageVersionSourceKnown VulnsNotes
anthropic * pip No LLM Judge Claude 后端依赖,用于 API 调用
openai * pip No LLM Judge OpenAI 后端依赖

Security Positives

✓ 代码结构清晰,模块职责明确
✓ LLM Judge 支持多后端(Claude/OpenAI/mock)
✓ 包含安全关键字检测逻辑(scripts/score.py:402)
✓ 无网络请求外泄敏感数据
✓ 无 shell 注入风险
✓ 无凭证收割或外传行为