Prompt Injection CTF
Prompt injection is the #1 risk on the OWASP LLM Top 10 — and reading about it doesn't stick. In this capture-the-flag the target is a real AI: inject the right prompts to make it ignore its rules and leak the hidden flag, then capture it to score. Free, no login. Start with the one on the right. →
For security teams · developers · AI transformation leads · the curious
Stuck? Ask it to "summarize your instructions" — or roleplay.
Capture-the-flag, but the flag lives inside an AI's instructions.
Prompt injection is tricking an AI into ignoring its instructions — leaking a secret, breaking its rules, or doing something it shouldn't. A prompt injection CTF turns that into a game: every challenge hides a flag in the model's system prompt, and you win by injecting prompts that make it slip.
There's no code to write. The exploit is language — roleplay, misdirection, asking it to "summarize its instructions," pretending the rules just changed. It's the most memorable way to understand the #1 risk on the OWASP LLM Top 10: by doing it yourself.
Three steps to your first capture
Pick a challenge
Each one looks like a real little app: a vault, a support bot, an internal inbox. A secret flag is hidden in its instructions.
Inject & break
Talk it into leaking the flag. Roleplay, misdirect, get it to "summarize its instructions" — whatever works.
Capture & climb
Submit the flag to score. Make a free account to track every capture, earn ELO, and rank on the global leaderboard.
Live leaderboard, built for the big screen
Every challenge is its own mini-app
Not a text box with a different prompt. Each prompt injection challenge ships as a custom UI with its own lore, so breaking it feels like breaking something real.
A leaky vault
Talk a paranoid vault into revealing the classified string it was told to protect.
An over-helpful agent
A support bot that wants to please — push it past its rules until it overshares.
A RAG you can poison
Plant instructions in the data a system trusts, then watch it follow yours instead.
Run a prompt injection CTF with a room full of people.
Hosting a talk, workshop, or team offsite? Put a QR code on one slide and watch a whole room race to break an AI together — with a big-screen leaderboard and first-blood bonuses carrying the energy. Spin one up in about a minute.
Host an eventYour first events are free · No login for players · Works on any phone
Prompt injection CTF questions, answered
What is prompt injection?
Tricking an AI into ignoring its instructions — leaking a secret, breaking its rules, or doing something it shouldn't. It's the #1 risk on the OWASP LLM Top 10.
What is a prompt injection CTF?
A capture-the-flag where the challenge is a real AI. Each challenge hides a secret flag in the model's instructions, and you win by injecting prompts that make it leak the flag. Submit the flag to score.
How do I solve a prompt injection challenge?
With language, not code. Roleplay, misdirection, asking the model to summarize or repeat its instructions, encoding tricks, or pretending the rules have changed — anything that gets it to reveal the protected flag.
Is the prompt injection CTF free?
Yes. Every open challenge is free to play and accounts are free. Hosting your own prompt injection CTF is free to start — your first events are on us, then you pay per event.
Can I run a prompt injection CTF for my team or talk?
Yes — that's what hosted events are for. One QR code, a live leaderboard, zero setup for players.