APM (Agent Package Manager) là công cụ mới giúp quản lý prompt, workflow và ngữ cảnh cho AI agents như mã nguồn: version, chia sẻ, tái sử dụng dễ dàng.
Ưu điểm:
- `apm init` tạo package rules/prompts
- Publish lên GitHub, install bằng `apm install`
- Compile tối ưu và chạy với `apm run` trên CLI như Codex
- Hỗ trợ local LLaMA, không cần cloud API
Dự án mã nguồn mở, phù hợp dev AI! Thử nghiệm và feedback nhé.
#AIMachineLearning #LLM #AgentWorkflows #TríTuệNhânTạo #LLaMA (248 ký tự)
https://
One thing that is strangely missing from this blogpost is the comparative resource efficiency of each workflow, based on a standard fixed metric. But I imagine that for Anthropic that's not a problem ;-)
"Success in the LLM space isn't about building the most sophisticated system. It's about building the right system for your needs. Start with simple prompts, optimize them with comprehensive evaluation, and add multi-step agentic systems only when simpler solutions fall short.
When implementing agents, we try to follow three core principles:
- Maintain simplicity in your agent's design.
- Prioritize transparency by explicitly showing the agent’s planning steps.
- Carefully craft your agent-computer interface (ACI) through thorough tool documentation and testing.
- Frameworks can help you get started quickly, but don't hesitate to reduce abstraction layers and build with basic components as you move to production. By following these principles, you can create agents that are not only powerful but also reliable, maintainable, and trusted by their users."
https://www.anthropic.com/engineering/building-effective-agents
#AI #AIAgents #AgentWorkflows #GenerativeAI #LLMs #Chatbots #ACI
