Study how popular AI coding assistants are prompted and configured internally.
Learn prompt engineering techniques by examining real-world examples from production AI tools.
Research security implications of exposed system prompts in AI products.
Based on the description and topics, this repository is a collection of system prompts, internal tools, and AI models extracted from a long list of named AI products. The description lists Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia, and v0, plus other open-sourced ones. The topics list adds bolt, cluely, copilot, cursor, cursorai, devin, and github-copilot. The README itself does not describe the contents in any structural way; it consists mainly of sponsor banners, support links, a security notice for AI startups about exposed prompts being attacked, contact information, and a star-history badge. As a result, the README does not provide further detail about how the prompts are organized, where they came from, what format they use, or how someone is meant to consume them. A complete explanation of the repository's structure and intended use is not possible from the provided data alone.
Generated 2026-05-18 · Model: sonnet-4-6 · Verify against the repo before relying on details.