Skip to content
Agentic Levels
Start hereLevelsTracksPromptsAbout
Get my levelGet my level

  • Start hereYour first useful task with AI.
  • Level quiz8 questions. No AI needed.
  • Fluency ScoreYour Level and a score, from your own AI.

  • LevelsThe 11 levels, in order.
  • LessonsShort lessons, filter by level.
  • TracksGuided paths by tool, role or goal.
  • Set up your AI tool
  • Starter packFree files that help AI remember.

  • PromptsReady prompts to copy and adapt.
  • ChatGPT vs Claude vs GeminiPick the right one for the job.
  • How-tos
  • ReferenceCommands for Claude Code, Codex, Antigravity.
  • ResourcesHand-picked reading and tools.

  • Your progress
  • Your next moveAn interview that picks your next skill.
  • Your next lessonsThe next lessons at your level.
  • Updates

  • Pricing
  • About
  • FAQ & Feedback
  • Work with us
  • How we handle your data
Sign inPreferences
AboutPricingHow-tosPromptsStarter packMy stackHow we handle your data

© 2026 Fuentes Studio

PrivacyTermsRefundsAccessibilityAI content policyCredits
yourCouncil
Ready to help
✦

What do you want to understand?

Ask anything about what you're learning.

←How-to guides
  1. Resources
  2. ›How-to guides

How to use AI System Prompts Collection

The most comprehensive collection of AI tool system prompts and model configurations, with security research on prompt injection risks.

Levels 2 to 4On GitHub

You're designing a system prompt or CLAUDE.md and want to see how production AI tools instruct their agents. This collection covers system prompts AND model configurations from major AI tools: Cursor, Devin, Bolt, Claude, and others. It also includes security guidance for AI developers on protecting against prompt injection. Community-contributed and regularly updated.

Before you start

No installation required, browse the repo

  1. 1Browse prompts for tools you recognize
  2. 2Apply the patterns to your own system prompt

Steps

01

Browse prompts for tools you recognize

Read the system prompts for tools you use yourself: Cursor, Claude Code, Copilot, Bolt. Compare what they emphasize: tool use guidelines, refusal patterns, output format rules, persona framing.

Analyze and extract patterns
I've read through the system prompts in the AI System Prompts collection, specifically [TOOLS_READ]. What are the common structural patterns across them? What do the best ones have that the others don't? Extract 5 patterns I should apply to my own CLAUDE.md or agent system prompt.
[TOOLS_READ]Which prompts you read e.g. Cursor, Claude Code, Bolt, Devin
02

Apply the patterns to your own system prompt

Pick 3-5 patterns from your analysis and rewrite your CLAUDE.md or system prompt to incorporate them. Test in a new session where Claude has no prior context.

How do you know it worked?

  • You've read at least 3 system prompts from tools you know
  • You've identified 3+ patterns you hadn't been using
  • Your own system prompt has been updated with at least 2 improvements

If something looks off

Treat as inspiration, not source of truth
Some prompts in the collection may be outdated or reconstructed. Verify against current tool behavior before relying on them.
Don't copy wholesale
Your context is different from Cursor's or Devin's, adapt patterns to your situation, don't paste blindly.

See also

  • Claude Code best practices
  • Awesome Cursor Rules

What you'll do, 2 steps

  1. 1Browse prompts for tools you recognize
  2. 2Apply the patterns to your own system prompt

Was this helpful?