This is a solid reference for building agents that can actually write code and use tools without going off the rails. You get the core agent loop (think, decide, act, observe), tool design patterns with proper schemas, and a thoughtful edit_file implementation that uses search/replace with occurrence counting to avoid clobbering the wrong lines. The multi-model architecture is smart: use fast models for planning, powerful ones for reasoning, specialized ones for code. What I appreciate is the focus on reliability, like the edit tool's conflict detection and the structured ToolResult pattern. If you're building anything that needs to autonomously modify codebases or chain tool calls, this gives you the architectural bones to do it safely.
npx -y skills add sickn33/antigravity-awesome-skills --skill autonomous-agent-patterns --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
sickn33/antigravity-awesome-skills
moizibnyousaf/ai-agent-skills
github/awesome-copilot