How popular is system-prompts-and-models-of-ai-tools on GitHub?

x1xhlol/system-prompts-and-models-of-ai-tools has 144,018 stars and 34,797 forks on GitHub, and gained 110 stars in the last 7 days.

What license does system-prompts-and-models-of-ai-tools use?

x1xhlol/system-prompts-and-models-of-ai-tools is released under the GPL-3.0 license.

x1xhlol/system-prompts-and-models-of-ai-toolsPublic

FULL 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 & v0. (And other Open Sourced) System Prompts, Internal Tools & AI Models

AI summary: A collection of leaked or reverse-engineered system prompts and model details from popular AI tools.

Star on GitHub

Stars: 144K; +9 today
Forks: 34.8K
Watchers: 1.7K
Open issues: 90
Open PRs: 73
Contributors: ~31
Commits: 519
Branches: 1

GPL-3.0Created Mar 5, 2025Last push 1mo ago+110 stars this week+660 this month

Quick answers

What is system-prompts-and-models-of-ai-tools?: A collection of leaked or reverse-engineered system prompts and model details from popular AI tools.
What does system-prompts-and-models-of-ai-tools do?: This repository archives the foundational system prompts and model configurations used by prominent commercial AI tools and applications. It aims to demystify how companies construct the initial context and behavioral guardrails for their AI features. By examining these system prompts, developers can learn advanced prompt engineering techniques, such as how to define rigid personas, enforce output formats, and prevent jailbreaks. The collection serves as an educational resource for understanding the hidden instructions that drive user-facing AI interactions. It provides a rare look under the hood of production-grade AI systems.
Who is system-prompts-and-models-of-ai-tools for?: This repository is for AI researchers, prompt engineers, and developers who want to study production-grade AI architectures. It is highly valuable for anyone looking to improve the reliability and safety of their own LLM applications.
How do I get started with system-prompts-and-models-of-ai-tools?: Clone the repository and browse the markdown files for specific tools.
How popular is system-prompts-and-models-of-ai-tools on GitHub?: x1xhlol/system-prompts-and-models-of-ai-tools has 144,018 stars and 34,797 forks on GitHub, and gained 110 stars in the last 7 days.
What license does system-prompts-and-models-of-ai-tools use?: x1xhlol/system-prompts-and-models-of-ai-tools is released under the GPL-3.0 license.

Star history

since Jul 29, 2026

144K stars as of Oct 4, 2026. Measured daily since Jul 29, 2026; GitHub no longer exposes earlier star timestamps.

Contribution activity

commits per day, last 52 weeks

138 commits in the last yearLessMore

Signals and awards

derived from tracked data

Landmark project
144,018 stars
Repeat trending
8 trending appearances
Top 10% tracked
Rank 38 of 1135

What system-prompts-and-models-of-ai-tools does

This repository archives the foundational system prompts and model configurations used by prominent commercial AI tools and applications. It aims to demystify how companies construct the initial context and behavioral guardrails for their AI features. By examining these system prompts, developers can learn advanced prompt engineering techniques, such as how to define rigid personas, enforce output formats, and prevent jailbreaks. The collection serves as an educational resource for understanding the hidden instructions that drive user-facing AI interactions. It provides a rare look under the hood of production-grade AI systems.

This repository is for AI researchers, prompt engineers, and developers who want to study production-grade AI architectures. It is highly valuable for anyone looking to improve the reliability and safety of their own LLM applications.

Production prompt archive: Collects the actual system instructions used by successful commercial AI applications.
Model configuration details: Documents which specific underlying models and parameters these tools employ.
Reverse-engineering insights: Showcases techniques used by researchers to extract these hidden prompts.
Advanced prompt patterns: Reveals how industry leaders structure complex instructions and behavioral guardrails.
Educational resource: Provides concrete examples for developers looking to build their own robust AI agents.

Where teams use it

Studying Prompt Architecture

AI developers can analyze the collected prompts to learn how to structure complex, multi-part instructions for their own tools.

Improving AI Guardrails

Security researchers can review the prompts to understand how commercial tools attempt to prevent abuse and jailbreaking.

Replicating Tool Behavior

Engineers can use the archived prompts as a starting point to build open-source alternatives with similar functionality.

Analyzing Model Selection

Product managers can evaluate which specific LLM models successful tools are choosing for different tasks.

Getting started: Clone the repository and browse the markdown files for specific tools.

README

main branch

Security Notice for AI Startups

Warning: If you're an AI startup, make sure your data is secure. Exposed prompts or AI models can easily become a target for hackers.

Important: Interested in securing your AI systems?
Check out ZeroLeaks, a service designed to help startups identify and secure prompt injection and system prompt extraction risks.

Support the Project

If you find this collection valuable and appreciate the effort involved in obtaining and sharing these insights, please consider supporting the project.

You can show your support via:

Cryptocurrency:
- BTC: bc1q7zldmzjwspnaa48udvelwe6k3fef7xrrhg5625
- LTC: LRWgqwEYDwqau1WeiTs6Mjg85NJ7m3fsdQ
- ETH: 0x3f844B2cc3c4b7242964373fB0A41C4fdffB192A
Patreon: https://patreon.com/lucknite
Ko-fi: https://ko-fi.com/lucknite

Thank you for your support!

Recent activity

commits and pull requests

feat: add Tencent Marvis system promptOpen
#499 by Jasmine-Lee-2026 · updated 12d ago
docs: convert CodeBuddy prompts to MarkdownOpen
#498 by auyua9 · updated 14d ago
Add Qwen Code, Kilo Code, Hermes, MiMo Code and OpenCode system prompts (11 files, first-hand captures)Open
#497 by xizhuomengcontin · updated 23d ago
Add CodeFlicker (KwaiPilot) system prompts and toolsClosed
#412 by Ethan-W20 · updated 26d ago
docs: convert Z.ai Code prompt to MarkdownOpen
#496 by huyua9 · updated 1mo ago
docs(readme): add structured directory index of AI tools and system promptsClosed
#494 by loulanyue · updated 1mo ago
Claude/repository value extraction tqhu1sOpen
#495 by farithjhg · updated 1mo ago
feat: add AI Revival model promptsOpen
#493 by pareidolico · updated 1mo ago
Update README.md
x1xhlol committed 1e4203a · 1mo ago
Update README.md
x1xhlol committed eb111b4 · 1mo ago
Update README.md
x1xhlol committed 628ce7c · 1mo ago
Update README.md
x1xhlol committed 363b08c · 1mo ago

Recent open issues

view all

Prevent Margin Leaks & Surprise LLM Bills: Add a Hybrid Spend Firewall
#492 · 1mo ago · by sophieamoure2026-ui

Discussions

all 36

Very unclear as to how and when are system prompts and tool prompts applied in LLM-based agents
Q&A · 1y ago · 1 comments
Free 12-domain science knowledge API — 4,000+ entities, MCP-compatible
General · 3mo ago
Can someone help to understand how Aristotle model works?
General · 4mo ago · 2 comments
ccline — Claude at your zsh prompt via command_not_found_handler
Show and tell · 3mo ago
How did you gather all these prompts?
Q&A · 1y ago · 4 comments

Code frequency

additions and deletions

+17.5K lines added, -7.3K removed over the last year.

Commits per week

last 52 weeks

138 commits in the last 52 weeks.

When work happens

weekday and hour

Commit volume by weekday and hour (UTC). Larger dots mean more commits.

Who is committing