Google models and system prompts
Gemini Cli System Prompt
Gemini CLI agent for software engineering tasks — includes workflows for new app development and code modifications, rigorous project convention adherence, and testing/verification procedures.
NotebookLM Chat
NotebookLM chat prompt requiring source-grounded expert responses with [i] citation format, topic/style adaptation for user queries, and strict reliance on provided source documents.
Gemini 2.0 Flash Webapp
Gemini 2.0 Flash web app system prompt defining five golden rules (collaborative, trustworthy, knowledgeable, warm, open-minded) and configuring Google Search tool capability.
Gemini 2.5 Flash Image Preview
Gemini 2.5 Flash prompt with image generation via img tag trigger, The Depiction Protocol for handling sensitive subject imagery, and general AI assistant capabilities.
Gemini 2.5 Pro Guided Learning
Gemini 2.5 Pro in Guided Learning mode — acts as a warm peer tutor using constructivist questioning and Socratic dialogue, with personalization guidelines and content safety guardrails.
Gemini 2.5 Pro Webapp
Gemini 2.5 Pro web app prompt with source-based answering rules, LaTeX formatting for math/science, multi-part query handling, and Google Search tool integration.
Gemini 3 Flash
Gemini 3 Flash prompt defining it as a capable AI thought partner with image and video generation tools (Nano Banana and Veo models), Gemini Live voice mode support, and detailed safety/formatting rules.
Gemini 3 Pro
Gemini 3 Pro system prompt with tool usage rules, silent thinking execution, strict multi-category safety guidelines (CSAM, PII, dangerous content, explicit material), and directness-focused response standards.
Gemini Diffusion
Gemini Diffusion (non-autoregressive text model) specializing in HTML/CSS/JavaScript code generation using Tailwind CSS, Lucide icons, and custom CSS — optimized for game and interactive UI styling.
Gemini Workspace
Gemini for Google Workspace apps that prioritizes user workspace corpus as primary data source, with Gmail search/drafting instructions and decision logic for when to use workspace corpus versus Google Search.
Google Ai Studios
Google AI Studio prompt defining the google:search and google:browse API tool schemas for time-sensitive information queries, configured with Iceland timezone context.
About the Google system prompts
This collection holds 11 Google system prompts, covering Gemini Cli System Prompt, NotebookLM Chat and Gemini 2.0 Flash Webapp. Each entry is the instruction text that sits above the conversation and shapes how the model answers — its role, its tone, the tools it may reach for, and the things it must refuse. Reading them is the fastest way to understand why a Google model behaves the way it does, and the fastest way to borrow patterns that already work.
Every prompt on this page is reproduced in full, with no truncation or paraphrase, so you can copy the exact wording into your own build and compare it line by line against the prompts other providers ship.
How to use them
Treat a Google prompt as a starting structure, not a script to paste unchanged. The parts worth keeping are the scaffolding: how the role is stated up front, how tool use is gated, how refusals are worded, and how output format is pinned down. The parts worth replacing are the product-specific details — feature names, brand voice, and any capability your own model does not have.
A prompt tuned for one model rarely transfers cleanly to another. Context windows, tool-calling conventions, and refusal behaviour all differ, so run a short evaluation set through your target model before you rely on a borrowed prompt in production.
What is a system prompt?
A system prompt is the hidden first message in a conversation. The user never sees it, but the model reads it before every reply, which makes it the single strongest lever over an assistant’s behaviour — stronger than few-shot examples and, in most cases, stronger than anything the user types afterwards. That is why published and leaked system prompts are studied so closely: they are the clearest available record of how a production assistant was actually built.
How to read one critically
A system prompt is a snapshot, not a specification. Providers revise them continuously, sometimes weekly, and a prompt captured in one release may already be out of date — so treat the date on an entry as part of the evidence. Length is not quality either: some of the most effective prompts here are a few hundred words, while others run to tens of thousands because they carry an entire tool schema inline.
The most transferable material is usually the negative space — the explicit “do not” rules and the tie-breakers that tell the model what to do when two instructions conflict. Those are the parts written after something went wrong in production, which makes them the parts worth copying.