Agent instructions
2026 04 14 Langchain context engineering
Core directives and parameters used to define the persona, constraints, and operational behavior of an AI Agent.
Context Engineering
The practice of optimizing the context-window by strategically populating it with precise information at each stage of an agent’s trajectory.
- Essential Context Components:
- Instructions: Primary directives for task execution.
- External Knowledge: Retrieved data (e.g., RAG).
- tool-feedback: Observational data from Tool Use/environment interactions.
- Implementation: Frameworks like LangGraph provide the infrastructure to manage these injection strategies and agent state.
Sources:
Source Notes
- 2026-04-07: Agent Skills Why Code Enhances LLM Efficiency Over Markdown for Scrapi · ▶ source
- 2026-04-08: Claude Cowork Desktop AI Co worker Core Capabilities and Advantages · ▶ source
- 2026-04-10: OpenClaw The Autonomous AI Agents Rise and Critical Security Flaws · ▶ source
- 2026-04-22: AI Agent Skills · ▶ source
- 2026-04-26: Karpathy
- 2026-04-27: Claude AI · ▶ source
- 2026-04-28: ChatGPT · ▶ source
- 2026-04-29: Hermes · ▶ source
- 2026-05-01: Modern AI Agentic Harness: Architecture, Components, and Framework Differences · ▶ source