forceinjection.github.io CLAUDE.md

Project-level instructions for Claude Code working in a Chinese-language Markdown knowledge repository about AI infrastructure, from hardware to AI applications.

In plain words
What is it for?
They guide Markdown content creation, directory placement, Conventional Commits, and coordination with the repository’s AGENTS.md instructions.
Why use it?
They explain the repository structure and rules so edits, content, and commit messages fit the project.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/forceinjection/forceinjection.github.io/claude-md
Clone the repo
git clone --depth 1 https://github.com/ForceInjection/forceinjection.github.io
Per session 2,172 This file is loaded in full into every session.
When invoked 2,172 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.02172 $0.02172
Opus 5 $0.01086 $0.01086
Sonnet 5 $0.00434 $0.00434
Haiku 4.5 $0.00217 $0.00217

Measured yesterday against content hash 326eb8767b4e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

forceinjection.github.io CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to AI-fundamentals CLAUDE.md — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

CLAUDE.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository overview

AI Fundamentals is a Chinese-language knowledge repository covering the full AI infrastructure stack: GPU architecture, CUDA programming, LLM theory, inference systems, cloud-native AI platforms, agentic systems, RAG, and more. All content is authored in Markdown.

  • License: Apache 2.0
  • Content is organized in semantically numbered top-level directories (01_hardware_architecture/ through 11_ai_native_everything/, plus 98_llm_programming/ and 99_misc/). Each directory corresponds to a major topic area with its own README.md portal.
  • 02_dpu_programming/02_gpu_programming/02_npu_programming/ share the 02_ prefix — all three are sub-modules under "底层计算与异构编程."
  • AGENTS.md exists alongside this file and covers module-level architecture details for GitHub Copilot. This file focuses on project-level conventions that apply to all work in the repo.

Commit conventions

This repo uses Conventional Commits with Chinese descriptions:

docs(scope): description
chore(scope): description
refactor(scope): description
feat(scope): description

Scopes are derived from directory/topic areas. Common scopes seen in the history: readme, dpu, gpu, npu, training, llm-theory, rag, graph_rag, agentic, agent_infra, inference, kv_cache, vllm, reference_design, storage, gpu_manager, k8s, course, trae, multi_agent, ai-native, smart_customer_service, kvbm, submodule. Look at git log --oneline for recent examples before committing.

File conventions

  • All top-level topic directories use zero-padded numeric prefixes (e.g., 01_, 02_, 03_) to maintain ordering.
  • Within a topic directory, files may use numeric prefixes for ordering (e.g., 01_concepts.md, 02_practice.md).
  • Translated content appends a language suffix to the filename (e.g., file.zh-CN.md).
  • Image assets live in img/ at the repo root, or alongside the files that reference them within topic subdirectories.
  • Interactive HTML visualizations (e.g., inference pipeline demos) are placed alongside the markdown documents they complement, in the same topic subdirectory.
  • README.md files at directory roots serve as navigation portals and contain link trees to content within that directory. When adding a new article, you must update the corresponding directory's README.md portal to include a link to the new file — this is the primary navigation mechanism for readers.

Read the full file on GitHub · 132 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 132 lines · 2,172 tokens per session scan A 326eb8767b4e

Subscribe to this mod's changes

forceinjection.github.io CLAUDE.md is an instructions file published in the GitHub repository ForceInjection/forceinjection.github.io (51 stars, last pushed 7d ago), licensed Apache-2.0. It adds 2,172 tokens to every session, about $0.0109 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to AI-fundamentals CLAUDE.md, differing in 0 lines, and is treated as a copy.