Tag:prompt engineering

Articles tagged "prompt engineering", page 2.

• 15 MIN READ•Mar 7, 2026

  Windsurf Flows vs MCP Servers: Context and MCP Config

  Windsurf Flows track your edits and terminal; MCP servers add outside tools. In Devin Desktop (ex-Windsurf), MCP lives in ~/.config/devin/mcp_config.json.

  AI toolscontext managementprompt engineering
• 14 MIN READ•Mar 7, 2026

  Zed AI Context: Agent Panel, AGENTS.md, Rules, and MCP

  Zed's Agent Panel loads one project instruction file (.rules first, then AGENTS.md or CLAUDE.md), takes @-mentions as context, and adds MCP via context_servers.

  AI toolscontext managementprompt engineering
• 13 MIN READ•Mar 7, 2026

  T3 Chat Guide: Models, $8 Plan, and Usage Limits

  T3 Chat puts GPT-6, Claude, Gemini, and Kimi behind one picker for $8 a month. Long threads, files, and search cost more, so here is how to keep context lean.

  AI toolscontext managementprompt engineering
• 15 MIN READ•Mar 7, 2026

  VS Code AI Context: Copilot, AGENTS.md, MCP, Cline, Continue

  VS Code Copilot reads copilot-instructions.md and AGENTS.md, loads MCP from .vscode/mcp.json, and compacts long chats. Cline and Continue add their own rules.

  AI toolscontext managementprompt engineering
• 21 MIN READ•Mar 7, 2026

  What Is MCP? Model Context Protocol and How Tools Connect

  MCP is an open protocol that lets AI apps call tools and read data on servers over stdio or Streamable HTTP. How it works and how 17 AI tools connect to it.

  AI toolscontext managementprompt engineering
• 14 MIN READ•Mar 7, 2026

  Codex Context Window: Limits, Compaction, and AGENTS.md

  Codex context window: GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna list 1.05M tokens in the API. Check your Codex limit with /status and free space with /compact.

  AI toolscontext managementprompt engineering