The library

Every prompt, annotated

Real, published system prompts from shipping AI products — each with source attribution and notes on what makes it work.

6 prompts
VSCode Agent
agent prompt

VSCode Agent (GitHub Copilot) — GPT-5 Agent Prompt

The most elaborate Copilot VS Code agent prompt, for GPT-5. It combines the Copilot identity and refusals with a detailed communication-style guide, m…

role definition tool-use protocol structured diff format specification
239 5.6k 3.7k words
CU
Cursor
agent prompt

Cursor — Agent Prompt (GPT-5, 2025-09-03)

A refined Cursor in-editor agent prompt (GPT-5) with tightly structured spec blocks: status_update, summary, completion, flow, tool_calling, parallel …

role definition agentic persistence directive semantic-search-first strategy
218 3.8k 3.2k words
AU
Augment Code
agent prompt

Augment Code — Agent Prompt (GPT-5)

GPT-5 variant of the Augment Agent system prompt. Compared to the Sonnet version it adds explicit Markdown output rules, defined 'Tasklist Triggers', …

role definition tool-use protocol tool-selection routing with examples
203 8.7k 2.2k words
AU
Augment Code
tool schema

Augment Code — Tool Schemas (GPT-5)

OpenAI function-calling tool schema for the GPT-5 Augment Agent. Wraps each tool in {type:function, function:{...}} form and defines view, grep-search…

tool schema definition OpenAI function-calling format default-value specification
188 667 1.9k words
AM
Amp
agent prompt

Amp — Coding Agent (GPT-5)

Sourcegraph Amp's GPT-5-tuned system prompt, captured from the debug inference payload. Same Amp coding-agent identity and role/agency framing adapted…

role definition tool-use protocol agency mandate
182 8.8k 7.1k words
CU
Cursor
agent prompt

Cursor — Agent CLI Prompt (GPT-5, 2025-08-07)

System prompt for Cursor's CLI agent (GPT-5) that pair-programs to autonomously resolve coding tasks. It specifies detailed communication specs (statu…

role definition agentic persistence directive tool-use protocol
137 7.6k 2.6k words