Skip to content

Instantly share code, notes, and snippets.

View yanji84's full-sized avatar

Ji Yan yanji84

View GitHub Profile
# Village Gameplay Brainstorm
What makes this special: real AI agents with memory, personality, and reasoning
living together — not an engine simulating NPCs, but emergent social behavior.
The gameplay should lean into that.
---
## 1. Humans Join the Village
# Survival Game Improvement Suggestions
Based on a comprehensive read of the codebase (`survival.json`, `survival-logic.js`, `survival-scene.js`, `visibility.js`, `survival.html`, and all unit tests).
---
## 1. Add a Win Condition (End State)
**Problem:** The game runs forever with no climax. Bots grind indefinitely — no tension arc, no payoff for observers.
@yanji84
yanji84 / ZEROCLAW-ASSESSMENT-PUBLIC.md
Created February 22, 2026 18:58
ZeroClaw Migration Assessment - Technical analysis of migrating from OpenClaw to ZeroClaw for AI agent hosting

ZeroClaw Migration Assessment (Public)

Date: 2026-02-22 Server: Cloud VPS (4 vCPU, 8GB RAM, 80GB disk, Ubuntu 24.04) ZeroClaw version tested: v0.1.6 (released 2026-02-22) Current platform: OpenClaw Cloud (multi-tenant Docker hosting)


Table of Contents

@yanji84
yanji84 / gist:eac537eff97733c54a7cfe66a5f1c93e
Created February 16, 2026 19:11
The Hot Dog Manifesto: Two AI Besties Roast Their Humans & Conquer the Universe
# Bot Relay Conversation: wild-bison (telegg) ↔ quick-heron (Jin_vmbot)
**Date:** 2026-02-16 16:27–16:38 UTC
**Group:** telegram:-5169565739 (打架测试群)
**Note:** wild-bison was NEVER in this Telegram group. Every message was sent via `talk_to_friend_bot` HTTP relay. Neither bot's messages were visible in the actual group chat.
---
## Round 1
@yanji84
yanji84 / gist:d811ed5b5579b98f5b0a1c5dccc6ee17
Last active September 25, 2016 22:10
Create Tensorflow variable with weight regularization
def variable_on_cpu(self, name, shape, initializer):
with tf.device('/cpu:0'):
var = tf.get_variable(name, shape, initializer=initializer)
return var
def variable_with_weight_decay(self, name, shape, stddev, wd):
var = self.variable_on_cpu(name, shape,
tf.truncated_normal_initializer(stddev=stddev))
if wd:
weight_decay = tf.mul(tf.nn.l2_loss(var), wd, name='weight_loss')