Trusted — Risk Score 0/100
Last scan:21 hr ago Rescan
0 /100
chatbot-engine
智能对话引擎 - 多轮对话与意图识别
This is a legitimate chatbot engine with no security issues - all declared capabilities match the implementation with no hidden functionality detected.
Skill Namechatbot-engine
Duration39.8s
Enginepi
Safe to install
This skill is safe to use. No action required.
ResourceDeclaredInferredStatusEvidence
Filesystem READ READ ✓ Aligned scripts/chatbot.py:103-115 save/load session methods
Network READ READ ✓ Aligned scripts/llm_adapter.py:58-62 API key env access + line 95-102 OpenAI API calls
Shell NONE NONE No subprocess/os.system calls found
Environment NONE READ ✓ Aligned scripts/llm_adapter.py:23 reads API keys for legitimate LLM functionality
Skill Invoke NONE NONE No skill invocation patterns found
Clipboard NONE NONE No clipboard access found
Browser NONE NONE No browser automation found
Database NONE NONE No database operations found

File Tree

10 files · 30.4 KB · 1054 lines
Python 7f · 982L Markdown 2f · 67L Text 1f · 5L
├─ 📁 examples
│ └─ 🐍 basic_usage.py Python 147L · 3.5 KB
├─ 📁 scripts
│ ├─ 🐍 chatbot.py Python 124L · 3.7 KB
│ ├─ 🐍 dialogue_manager.py Python 71L · 2.4 KB
│ ├─ 🐍 intent_classifier.py Python 166L · 4.8 KB
│ ├─ 🐍 knowledge_base.py Python 221L · 6.4 KB
│ └─ 🐍 llm_adapter.py Python 156L · 5.1 KB
├─ 📁 tests
│ └─ 🐍 test_chatbot.py Python 97L · 2.8 KB
├─ 📝 README.md Markdown 9L · 96 B
├─ 📄 requirements.txt Text 5L · 78 B
└─ 📝 SKILL.md Markdown 58L · 1.7 KB

Dependencies 5 items

PackageVersionSourceKnown VulnsNotes
openai >=1.0.0 pip No Version lower bound set, minor version unpinned
scikit-learn >=1.3.0 pip No Version lower bound set, minor version unpinned
numpy >=1.24.0 pip No Standard ML dependency
pandas >=2.0.0 pip No Standard data processing dependency
regex >=2023.0.0 pip No Standard regex library

Security Positives

✓ No shell execution (subprocess, os.system, etc.) - clean codebase
✓ No network exfiltration - only legitimate API calls to declared providers (OpenAI, Anthropic)
✓ No credential harvesting beyond necessary API key access for LLM functionality
✓ No obfuscation techniques (base64, eval, etc.) detected
✓ All file operations are declared and documented (session save/load, config persistence)
✓ Dependencies are properly version-pinned in requirements.txt
✓ Plugin system is declared and properly scoped
✓ No sensitive path access (no ~/.ssh, ~/.aws, .env scanning)
✓ No persistence mechanisms (no cron, startup hooks, backdoors)
✓ Clean code with no hidden functionality