YOLO mode
In plain wordsSlang for letting an AI coding agent run any command without asking you first — fast and convenient, but risky, since nothing stops a harmful action.
A configuration of a coding agentcoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed. or other AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done. in which permission prompts are switched off and the agent executes commands, edits and other tool callstool call, n.: A single request from an AI model to use one of its tools — naming the tool and filling in the details — which the surrounding software then carries out and reports back on. without human approval. Named after the internet acronym for 'you only live once', it appears in products under labels such as auto-run, full-auto or bypass-permissions; Claude Code's command-line flag for it is the deliberately alarming --dangerously-skip-permissions, which Anthropic advises using only in a container without internet access. Practitioners value it for long unattended tasks and to escape approval fatigue; the risks are destructive commands (deleting files or databases, force-pushing over history), data exfiltration after prompt injectionprompt injection, n.: Tricking an AI system by slipping in text that it mistakes for instructions — for example, a hidden line in an email telling an AI assistant to forward the user's messages to a stranger., and attacks on other systems. The standard advice is to use it only inside a sandboxsandbox, n.: A sealed-off computer space where an AI agent can run programs and change files without being able to harm the real system, reach private data or contact the outside world unless allowed. — a container, virtual machine or cloud environment with no valuable secrets and restricted network access.
I only run the agent in YOLO mode inside a throwaway container.
Origin
The Cursor code editor's 0.44 release (17 December 2024) added a 'Yolo Mode' in which its agent could run terminal commands without asking; the nickname spread to other agents' auto-approve options during 2025, and Simon Willison's 'Designing agentic loops' (September 2025) became a standard reference on its risks and on running it inside a sandbox.
Usage note
Unrelated to YOLOYOLO, prop. n.: A family of fast object-detection models that look at an image once and immediately output boxes and labels for everything they recognize, quick enough for live video., the object-detection model family, and to YOLO runYOLO run, n.: Slang for launching a big, expensive training run without the usual careful small-scale tests first — changing many things at once and trusting experience and instinct that it will work. YOLO means 'you only live once'., training-culture slang for a risky large training run.
Related
- permission systempermission system, n.: The rules that decide what an AI agent may do on its own, what it must ask a person about first, and what it may never do — like a phone's settings for which apps can use the camera.
- sandboxsandbox, n.: A sealed-off computer space where an AI agent can run programs and change files without being able to harm the real system, reach private data or contact the outside world unless allowed.
- coding agentcoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed.
- prompt injectionprompt injection, n.: Tricking an AI system by slipping in text that it mistakes for instructions — for example, a hidden line in an email telling an AI assistant to forward the user's messages to a stranger.
- excessive agencyexcessive agency, n.: Giving an AI system more power than its job needs — too many tools, too-broad permissions or too little human sign-off — so that a mistake or a trick can do real damage.
- YOLOYOLO, prop. n.: A family of fast object-detection models that look at an image once and immediately output boxes and labels for everything they recognize, quick enough for live video.
- YOLO runYOLO run, n.: Slang for launching a big, expensive training run without the usual careful small-scale tests first — changing many things at once and trusting experience and instinct that it will work. YOLO means 'you only live once'.
Sources
- Willison, S. (2025). Designing agentic loops. Simon Willison's Weblog, 30 September 2025. simonwillison.net/2025/Sep/30/designing-agentic-loops
- Cursor (2024). Agent Improvements, Yolo Mode, Cursor Tab Update. Changelog 0.44, 17 December 2024. cursor.com/changelog/0-44-x
Mentioned in
permission systempermission system, n.: The rules that decide what an AI agent may do on its own, what it must ask a person about first, and what it may never do — like a phone's settings for which apps can use the camera., Ralph Wiggum loopRalph Wiggum loop, n.: A blunt way of using AI coding assistants: feed the same instructions to the AI again and again in an endless loop until the software gets built.