zensh

DeepSeek Harness 内测申请汇总与分析报告

• 帖子: https://x.com/tianyi/status/2083519855203078320 (Tianyi Cui, 2026-08-01 19:46 +08)
• 数据采集时间: 2026-08-02 08:00–08:30 UTC（发帖后约 20 小时）
• 采集方法: X 搜索 conversation_id:2083519855203078320 全量滚动（最新→最旧），去重后共 425 条回复，时间覆盖 2026-08-01 11:49 UTC（发帖后 3 分钟）至采集时点，完整覆盖整个对话。
• 口径说明: 帖子页面显示 697 条回复，搜索索引实际返回 425 条（含嵌套回复；部分回复可能未入搜索索引或被删除）。本报告基于可公开检索到的全部内容。

---

一、总体统计

karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

codeskyblue

# author: codeskyblue 2025
# https://github.com/openatx/uiautomator2
# Example: how to manually handle popup-windows

import random
import time
from typing import Optional, Sequence, Tuple

import adbutils
from loguru import logger

nkhitrov

"""
Structlog example configuration with FastAPI.
Features:
- async bound logger
- contextvars to log request-id and other meta data
- custom format for default logging loggers and structlog loggers
"""

import asyncio
import logging

porthunt

import base64
import json
import requests

REPO_URL = "https://api.github.com/repos/<USER>/<REPO>/contents/<PATH>/<TO>/<FILE>.json"
TOKEN = "<YOUR PAT OR OAUTH TOKEN>"

headers = {
    "Authorization": f"token {TOKEN}",
    "Accept": "application/vnd.github.v4+raw"

y0ngb1n

Docker Hub 镜像加速器

国内从 Docker Hub 拉取镜像有时会遇到困难，此时可以配置镜像加速器。

Dockerized 实践 https://github.com/y0ngb1n/dockerized

1️⃣ Docker daemon 配置代理（推荐）

参考 Docker daemon 配置代理

wavezhang

wget -c --no-cookies --no-check-certificate --header "Cookie: oraclelicense=accept-securebackup-cookie"  https://download.oracle.com/otn-pub/java/jdk/12.0.2+10/e482c34c86bd4bf8b56c0b35558996b9/jdk-12.0.2_linux-x64_bin.tar.gz

madrobby

curl -H 'Authorization: token INSERTACCESSTOKENHERE' -H 'Accept: application/vnd.github.v3.raw' -O -L https://api.github.com/repos/owner/repo/contents/path

jwebcat

wget --no-check-certificate --content-disposition https://github.com/joyent/node/tarball/v0.7.1
# --no-check-cerftificate was necessary for me to have wget not puke about https
curl -LJO https://github.com/joyent/node/tarball/v0.7.1