How System Prompts Define Agent Behavior

kept by eddie feb 17

System prompts shape coding agent behavior as much as the underlying model, with swaps between agents like Claude and Codex producing distinctly different workflows.

GitHub - es617/claude-replay: Community tool to convert Claude Code session transcripts into self-contained, embeddable HTML replays
Claude-replay converts AI coding agent session transcripts into self-contained, interactive HTML replays for sharing and embedding.
also on Gemini CLI, codex-cli, Cursor, Claude Code, #ai-coding, #developer-tools
Solo
Solo is a native terminal workspace that runs AI coding agents like Claude Code and Codex alongside a developer's full stack, providing unified visibility, auto-restart, and MCP-based agent coordination.
also on Gemini CLI, codex-cli, Claude Code, #ai-coding, #developer-tools
What you're actually writing when you write a SKILL.md
Skills are loader specifications with three execution stages, not static prompts, and understanding this runtime architecture is the key to writing efficient, portable agent skills.
also on codex-cli, Cursor, Claude Code, #ai-coding, #software-engineering, #developer-tools
Import from another agent
ChatGPT and Codex now support importing setup and recent work from Claude Code, Claude Cowork, or Cursor via a guided flow that preserves existing configurations and enables sync.
also on codex-cli, Cursor, Claude Code, #ai-coding, #developer-tools