Langchain Skill Vmisep 2026
Langchain Skill Vmisep 2026 sử dụng LangChain với memory và chain, hỗ trợ trả lời tiếng Việt, nhớ lịch sử chat và mở rộng RAG, tool calling, agent tự động.
MIT-0 · Free to use, modify, and redistribute. No attribution required.
⭐ 0 · 20 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The code implements a LangChain-based chat with memory and a routing decision between 'DeepSeek' and 'Gemini', which aligns with the description. However, the skill includes an embedded API key and a DeepSeek API base URL inside the code despite the registry metadata declaring no required env vars or credentials—this credential presence is disproportionate to the metadata and should have been expressed as a required credential or configurable environment variable.
Instruction Scope
SKILL.md instructs using LangChain and mentions editing api_base/api_key in langchain_skill.py but does not warn that the distributed code already contains a hard-coded secret. The runtime instructions and code will transmit queries (and chat history via memory) to external services (DeepSeek and Google Gen AI). That network behavior and embedded credential are not surfaced as required credentials in the skill metadata.
Install Mechanism
There is no install spec (instruction-only model with a Python file), which minimizes install-time risk. However the skill depends on third-party Python packages (langchain and provider-specific packages) as noted in SKILL.md; those must be installed in the agent environment before run and will cause network/API calls at runtime.
Credentials
The skill declares no required environment variables but the code contains a hard-coded API key (openai_api_key="sk-e7ec5...39506694") and a fixed api_base (https://api.deepseek.com/v1). This is inconsistent: credentials are embedded rather than requested via config/env, and the skill will send user queries and memory to third-party endpoints using that embedded key. That raises privacy, consent, and credential-management concerns.
Persistence & Privilege
The skill is not always-enabled and has no install-time persistence spec. It does not request system-level privileges or modify other skill configurations. Autonomous invocation is allowed (platform default) but is not combined with elevated installation privileges.
What to consider before installing
This skill behaves like a LangChain chat assistant but contains a hard-coded API key and endpoint inside langchain_skill.py that will be used to call external LLM services. Before installing or running: (1) Do not run it with sensitive data — your messages and memory will be sent to third-party APIs. (2) Treat the embedded key as a secret leak — it should be removed and replaced with a config-driven approach (use environment variables) or you should obtain your own key. (3) Verify the key owner and consider rotating any exposed keys. (4) If you trust the author and want to use it, edit the code to read credentials from env vars (and update the skill metadata to declare them), or remove the hard-coded key entirely. (5) Only install packages from trusted sources and review network endpoints the skill calls. The mismatch between declared metadata (no creds) and embedded credentials is the primary reason this skill is suspicious.Like a lobster shell, security has layers — review code before you run it.
Current versionv1.0.0
Download ziplatest
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
SKILL.md
LangChain Python Skill - Trợ lý thông minh với memory và chain
Skill này sử dụng thư viện LangChain để tạo chain LLM với bộ nhớ conversation buffer, giúp bot nhớ ngữ cảnh từ các tin nhắn trước trong cùng session. Skill hỗ trợ trả lời bằng tiếng Việt, ngắn gọn, chính xác, và có thể mở rộng để thêm RAG (PDF/document), tool calling (search, calculator), hoặc agent tự quyết định.
Tính năng chính
- Conversation memory: Nhớ lịch sử chat trong session (dùng ConversationBufferMemory).
- Prompt template tùy chỉnh: Có thể chỉnh sửa prompt để phù hợp với persona của bot.
- Backend LLM: DeepSeek-chat (hoặc Gemini nếu thay đổi key).
- Usage: langchain <query> hoặc langchain: <query>
- Ví dụ thực tế: langchain: Blockchain là gì? langchain: Tôi tên S0nSun
name: LangChain Test Skill của Sếp slug: langchain-skill-vmisep-2026 description: Skill test LangChain tích hợp bởi S0nSun & Grok version: 1.0.1
langchain: Tên tôi là gì? (bot sẽ nhớ và trả lời chính xác)
Hướng dẫn sử dụng nâng cao
- Để test memory: Chat nhiều câu liên tiếp với cùng từ khóa "langchain".
- Có thể mở rộng: Thêm RAG bằng Chroma vector store, tích hợp tool web search (Tavily), hoặc gọi API bên ngoài.
- Yêu cầu: Cần venv với langchain đã cài (pip install langchain langchain-community langchain-core).
Lưu ý
- Skill chạy trên Python, dùng OpenAI wrapper để gọi DeepSeek API.
- Nếu muốn dùng Gemini: Chỉnh api_base và api_key trong langchain_skill.py.
- Author: Tích hợp bởi S0nSun & Grok (dựa trên LangChain docs 2026)# LangChain Test Skill
- Description: Skill test LangChain đơn giản để kiểm tra load
- Usage: langchain <query>
- Example: langchain: Xin chào
Tối ưu chi phí mới nhất (2026)
- Router tự động: Query tiếng Việt/code → ưu tiên DeepSeek (rẻ hơn).
- Memory summary: Tự tóm tắt lịch sử chat để giảm input tokens.
- Prompt cache: Giảm lặp lại system prompt.
- Output giới hạn: Trả lời ngắn gọn dưới 200 từ.
Files
2 totalSelect a file
Select a file to preview.
Comments
Loading comments…
