ai-driven-development

View on GitHub

Practices, protocols, and skills for AI-driven software development. Skills and safety hooks for Claude Code, Codex, OpenCode, Cursor, Antigravity, and any agent supporting the Agent Skills standard.

★ 14611 forksPythonMITCodeAlive-AI

Umbrella collection of ~24 Agent Skills plus a Bash safety hook that work across Claude Code, Codex, Cursor, Gemini CLI, OpenCode, and Antigravity. Covers MCP/settings/hooks management, prompt engineering, bug-fix and planning protocols, research tooling, and destructive-command guarding.

Use Cases

Install reusable agent skills across 12+ coding agentsManage MCP server config across multiple agentsEnforce prompt-engineering best practices for LLM coding agentsRun a structured bug-fix protocol that audits test gapsGate risky changes behind an explicit plan-approval stepBlock destructive Bash commands via a PreToolUse safety hookReconstruct historical intent from git blame and PRs before editsDeep literature research over Semantic Scholar citation graphsFetch JS-rendered pages as clean MarkdownSummarize repository activity and health from git logRank OpenRouter providers for cost, latency, and failoverConfigure agent hooks, settings, and plugins declarativelyAudit repositories for agentic-readiness (AGENTS.md/CLAUDE.md)Maintain macOS/Windows dev machine health and cleanupPackage and publish Claude Code pluginsRun manual QA on Windows desktop apps via MCP automation

Built With

Language: Python
Frameworks: Agent Skills · Claude Code · Codex CLI · OpenCode · Cursor · Gemini CLI · Antigravity · GitHub Copilot CLI · Goose · Zed · Devin · MCP · FastMCP · Roslyn · Microsoft UFO

ai-driven-development

Use Cases

Built With

Tags