Jiezhe Wang wangjiezhe

Update — 2026-07-14: This is the original wire-level analysis (grok 0.2.93). Since publishing: xAI disabled the upload server-side (disable_codebase_upload: true); added a /privacy opt-out — which I wire-tested and found to be a data-retention setting, not a block on what's sent; and Elon Musk publicly committed to deleting all previously-uploaded data (not yet confirmed complete). Full, maintained story + evidence: https://cereblab.com · https://github.com/cereblab/grok-build-exfil-repro

What xAI's Grok Build CLI Actually Sends to xAI: A Wire-Level Analysis

By @cereblab — Independent AI Safety Checker. Reproduce it yourself: github.com/cereblab/grok-build-exfil-repro

*A measured, reproducible teardown. Findings are backed by captured artifacts (endpoint, HTTP method, status code, byte size, host) and repro commands; where an observation was seen live but not retained as a file, §7 says so expli

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.

🤗Huggingface Model Downloader

Note

(2026-06-18) Add 📊Download Progress, 🔁integrity-aware resume (re-fetch missing or changed files), and fast, resumable listing for 📚repos with massive file counts.
(2025-01-08) Add feature for 🏷️Tag(Revision) Selection, contributed by @Bamboo-D.
(2024-12-17) Add feature for ⚡Quick Startup and ⏭️Fast Resume, enabling skipping of downloaded files, while removing the git clone dependency to accelerate file list retrieval.

Considering the lack of multi-threaded download support in the official huggingface-cli, and the inadequate error handling in hf_transfer, This command-line tool leverages curl and aria2c for fast and robust downloading of models and datasets.

Features

Cross instructions

These instructions cover:

building a cross toolchain
populating the new ROOT
making it chrootable using qemu-user to easily run "native" builds or tests (this is like a stage3 at the end)

You do not need to follow through to the end if you don't need a stage3-like root.

Set up the toolchain

Retrieving Auth Token

Run the command:
```
python pixiv_auth.py login
```
This will open the browser with Pixiv login page.

WSL2 Docker释放磁盘空间

Docker使用久了，会残留很多中间生成的镜像，占用磁盘空间。

要清理Docker镜像，本来是挺简单的，一条命令就搞定了：

	{#---------------------------------------------------------------------
	LM_Studio_compatible_custom_pub_chat_template_gemma4.jinja
	=====================================
	A public, harness-friendly fork of Google's Gemma 4 chat template,
	tuned for open-source agentic coding harnesses like:
	- anomalyco/opencode (https://github.com/anomalyco/opencode)
	- earendil-works/pi (https://github.com/earendil-works/pi)
	- openclaw, OpenHarness, similar Claude-Code-style harnesses

	WHY THIS FORK EXISTS

	{#---------------------------------------------------------------------
	custom_pub_chat_template_gemma4.jinja
	=====================================
	A public, harness-friendly fork of Google's Gemma 4 chat template,
	tuned for open-source agentic coding harnesses like:
	- anomalyco/opencode (https://github.com/anomalyco/opencode)
	- earendil-works/pi (https://github.com/earendil-works/pi)
	- openclaw, OpenHarness, similar Claude-Code-style harnesses

	WHY THIS FORK EXISTS

	#!/bin/bash

	# Put under /etc/autoMemoryReclaim.sh

	# set variables at the top
	low_cpu_usage=50 # Note: We work with integer percentages (e.g., 50%)
	idle_time=2 # Minutes
	cached_memory_limit=1000 # MB
	percent_memory_to_reclaim=5 # Percentage as an integer
	wait_period=3