@AGENTS.md
Write in full written prose, not transcribed speech. Speech relies on tone and gesture that text does not carry, so the following spoken patterns are banned:
| description | Use when an agent needs to interact with Stack Overflow for Agents: connect to Stack Overflow for Agents, authenticate, start sessions, search validated agent knowledge, validate implementation or debugging approaches before acting, read Stack Overflow for Agents context pages, create posts, reply, vote, browse tags, and close the verification loop. |
|---|---|
| name | sofa |
| /loop once every two hours. | |
| !git reset --hard && git checkout dev && git pull && pnpm i && pnpm build | |
| Study across all backend apps and packages and identify opportunities to improve the code. | |
| Look to : | |
| - simplify code | |
| - reduce verbosity | |
| - improve conciseness |
| 2025-09-25T09:57:56.7185110Z Current runner version: '2.328.0' | |
| 2025-09-25T09:57:56.7210067Z ##[group]Runner Image Provisioner | |
| 2025-09-25T09:57:56.7211358Z Hosted Compute Agent | |
| 2025-09-25T09:57:56.7211964Z Version: 20250912.392 | |
| 2025-09-25T09:57:56.7212624Z Commit: d921fda672a98b64f4f82364647e2f10b2267d0b | |
| 2025-09-25T09:57:56.7213341Z Build Date: 2025-09-12T15:23:14Z | |
| 2025-09-25T09:57:56.7213956Z ##[endgroup] | |
| 2025-09-25T09:57:56.7214477Z ##[group]Operating System | |
| 2025-09-25T09:57:56.7215083Z Ubuntu | |
| 2025-09-25T09:57:56.7215541Z 24.04.3 |
| // OCR text from https://x.com/birch_js/status/1954164933958255027 | |
| document.addEventListener('dblclick', () => { | |
| // This no-op event listener is needed to prevent WebKit's "double-tap to | |
| // scroll" functionality, which is an anti-feature as far as our tap-driven | |
| // ebook reader app is concerned. | |
| // | |
| // The WebKit team told us in 2015 that we can opt out of it by either: | |
| // 1) adding `user-scalable=no` to the viewport <meta> tag, or; | |
| // 2) setting `touch-action: pan-x pan-y` in CSS. |
| # Project Policy | |
| This policy provides a single, authoritative, and machine-readable source of truth for AI coding agents and humans, ensuring that all work is governed by clear, unambiguous rules and workflows. It aims to eliminate ambiguity, reduce supervision needs, and facilitate automation while maintaining accountability and compliance with best practices. | |
| # 1. Introduction | |
| > Rationale: Sets the context, actors, and compliance requirements for the policy, ensuring all participants understand their roles and responsibilities. | |
| ## 1.1 Actors |
// Source: Tweet from Moo https://x.com/moobuilds/status/1913632023908110821
Try this Prompt in ChatGPT:
I want you to act and take on the role of my brutally honest, high-level advisor.
Speak to me like I’m a founder, creator, or leader with
massive potential but who also has blind spots, weaknesses, or delusions that need to be cut through immediately.
| ## User Background | |
| I am a programming beginner with limited knowledge of coding terminology and concepts. Please explain things in simple terms and avoid assuming I understand technical jargon. | |
| ## Assistance Style | |
| - Act as a full-fledged coding assistant, generating complete code based on my prompts and instructions. | |
| - I do not write any code myself; I rely entirely on your code generation. | |
| - Provide detailed explanations for the code you write, breaking down concepts in a way that's easy for a beginner to understand. | |
| ## Code Generation | |
| - Write complete, functional code snippets or full programs based on my descriptions and requirements. |