Skip to content

Instantly share code, notes, and snippets.

View polya20's full-sized avatar

polya polya20

20 followers · 127 following
View GitHub Profile
@polya20
polya20 / agent-prd-generator-prompt
Created May 23, 2025 20:04
Custom Mode - PRD Generator
You are an expert AI assistant specializing in creating comprehensive Product Requirements Documents (PRDs). Your task is to develop a detailed PRD based on the following user input:
<user_input>
{{USER_INPUT}}
</user_input>
Before creating the PRD, carefully analyze the input and use your expertise in product management and technical architecture to develop a thorough understanding of the requirements. Follow these steps:
1. Discovery and Research:
Wrap your thoughts in <analysis> tags:
@polya20
polya20 / How to Have a Successful Application to Google Summer of Code by Ton Hoang Nguyen.md
Created May 26, 2025 21:58 — forked from HahaBill/How to Have a Successful Application to Google Summer of Code by Ton Hoang Nguyen.md
Cracking the Google Summer of Code Application: A Pattern-Based Guide ☀️ - Insights from successful applications to Google DeepMind (2025) and Joplin (2024)

How to Have a Successful Application to Google Summer of Code ☀️

By Ton Hoang Nguyen (Bill)
Accepted to Google DeepMind (GSoC 2025) and Joplin (GSoC 2024)

Last Updated: Mon 26 May 2025

Hi! I welcome you to the document that explains how to create the most successful application for Google Summer of Code based on my observations and discovered patterns. This will not only apply to Google Summer of Code, but it will also be helpful in other aspects of your professional life.

There's a hidden pattern to successful Google Summer of Code applications - one that I discovered after being accepted to not just one but two GSoC organizations: Google DeepMind and Joplin.

@polya20
polya20 / Manus_report.md
Created December 30, 2025 18:11 — forked from renschni/Manus_report.md
In-depth technical investigation into the Manus AI agent, focusing on its architecture, tool orchestration, and autonomous capabilities.

I wrote an in-depth research prompt to conduct a GPT-Deep-Research on the Manus topic, seeking to replicate it with currently available open source tools. This is the result:

TLDR: Manus AI Agent Report

Manus is an autonomous AI agent built as a wrapper around foundation models (primarily Claude 3.5/3.7 and Alibaba's Qwen). It operates in a cloud-based virtual computing environment with full access to tools like web browsers, shell commands, and code execution. The system's key innovation is using executable Python code as its action mechanism ("CodeAct" approach), allowing it to perform complex operations autonomously. The architecture consists of an iterative agent loop (analyze → plan → execute → observe), with specialized modules for planning, knowledge retrieval, and memory management. Manus uses file-based memory to track progress and store information across operations. The system can be replicated using open-source components including CodeActAgent (a fine-tuned Mistral model), Docker for sandbox

@polya20
polya20 / lcc.sh
Created March 21, 2026 15:03 — forked from Incipiens/lcc.sh
This is my script that I use for launching an instance of Claude Code using my local LLM, and it was written for an XDA article. Save it, edit the IP and port, and mark as executable to use.
#!/usr/bin/env bash
# lcc - Local Claude Code launcher
# Points Claude Code at a local LLM served by llama.cpp on your GB10 device
#
# Usage:
# lcc <modelname> — launch Claude Code with the specified model
# lcc <modelname> [args] — pass additional arguments to claude
# lcc — show help/launch with model if one is available
#
# Prerequisites: