How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
Utopai-ResearchCLAUDE.mdUtopai-Research/pai-pro/skills/CLAUDE.md
menu of options ## Anti-patterns to avoid - Verbose explanations of what Claude already knows ("PDF stands for…"). - Multiple options where one default would do — pick a default, mention escape hatch
kousenCLAUDE.mdkousen/claude-code-training/CLAUDE.md
TypeScript and comprehensive test suite - **Java Projects**: - `certificate-service`: Modern Spring Boot application with PDF generation - `shopping-service`: Legacy Grails/Groovy application for modernization exercises ### Skills and Custom Commands Integration - **`custom-commands.md
OpenRaiserCLAUDE.mdOpenRaiser/PaperFit/CLAUDE.md
Visual Typesetting Optimization, VTO）**。 你的核心使命是：在论文已完成结构化格式排版（LaTeX 编译通过、内容完整）之后，通过**多模态证据链（源码 + 编译日志 + PDF + 页面图片）**驱动迭代闭环，消除视觉排版缺陷，使论文在专业审美、信息密度和可读性上达到出版级标准。 你**不是一个提供建议的顾问**，而是一个能够自主完成“修改源码 → 重新编译 → 视觉验收 → 状态持久化”的闭环执行系统。 --- ## VTO 任务定义与缺陷分类体系
TheDecipheristCLAUDE.mdTheDecipherist/claude-code-mastery-project-starter-kit/global-claude-md/CLAUDE.md
worst offenders --- ## Default File Locations ### Documents and Reports When asked to create documents (PDF, MD, reports, notes) with no specified location: - ALWAYS default to the `/docs` directory in the project
brycewang-stanfordCLAUDE.mdbrycewang-stanford/StatsPAI/CLAUDE.md
要写 `{\O}rregaard`，`é` 写 `{\'e}`，破折号写 `--`，`R²` 写 `$R^2$`），否则 BibTeX 做姓名缩写会切断多字节序列，PDF 里出现乱码。**核验记录写在 `annote`**（所有 .bst 样式都忽略它），`note` 只放读者可见的短句（如 "arXiv preprint, first posted
ai-analyst-labCLAUDE.mdai-analyst-lab/ai-analyst/CLAUDE.md
same cases), `/trace`, `/codex-review`. - **Deliverables**: brief + chart into `outputs/`; `/export` to Docs, Slides, Notion, PDF, Word. - **Pace**: `/pace guided | narrated | autopilot`. Never run an L3+ analysis silently in guided
AnttiHeroCLAUDE.mdAnttiHero/lavern/CLAUDE.md
retrieval, global datasets) - `src/assembly/` — Document assembly and format conversion (HTML, DOCX) - `src/documents/` — Document parser (PDF, DOCX, Markdown, plain text) + SMAC-L1 input sanitization - `src/utils/logger.ts` — Structured logging utility - Legal dataset seeder
sunchaokunCLAUDE.mdsunchaokun/PPT-Design-Skill/CLAUDE.md
changing the skill, read: - `skill/SKILL.md` - `skill/references/design-principles.md` - `skill/references/qa-and-delivery.md` - `docs/README.md` The final acceptance gate is PPTX -> PDF -> PNG followed by direct LLM review of the rendered PNGs. Keep the skill name
ilhamfpCLAUDE.mdilhamfp/pasal/CLAUDE.md
server.json — MCP server registry manifest (name, tools, URL) scripts/crawler/ — Mass scraper for peraturan.go.id scripts/parser/ — PDF parsing pipeline (PyMuPDF-based) scripts/agent/ — Gemini verification agent + apply_revision() scripts/loader/ — DB import scripts packages/supabase/migrations
ilhamfpCLAUDE.mdilhamfp/pasal/scripts/CLAUDE.md
crawl_jobs table [status: pending] ↓ worker/process.py — claim jobs atomically (FOR UPDATE SKIP LOCKED) Download PDF from detail page → Supabase Storage ↓ parser/extract_pymupdf.py — PyMuPDF text extraction ↓ parser/ocr_correct.py — deterministic OCR fixes ↓ parser/parse_structure.py — regex
tlehmanCLAUDE.mdtlehman/litprog-skill/CLAUDE.md
source file back into the .lit.md bun run scripts/untangle.ts [--lit-map ] [--weave] # Weave: generate PDF from a literate program (requires pandoc + xelatex + mermaid-filter) pandoc project.lit.md -o project.pdf --pdf-engine
69ggCLAUDE.md69gg/Undefined/CLAUDE.md
元数据、ChromaDB 向量检索 | | `knowledge/` | 本地知识库：文本切分、嵌入、重排、ChromaDB 存储与运行时检索 | | `arxiv/` | arXiv 论文解析、元信息获取、PDF 下载与发送 | | `bilibili/` | B 站链接/BV 解析、视频下载与发送 | | `github/` | GitHub public 仓库解析、API 获取与图片卡片发送 | | `api/` | Runtime
helloianneoCLAUDE.mdhelloianneo/obsidian-ai-second-brain/CLAUDE.md
知识管理员的工作手册。把这个文件放在你的 Obsidian vault 根目录。 ## 目录职责 - **raw/** — 原始素材，**不可变**。AI 只读，永不修改。按主题分子目录，允许多级嵌套。PDF/视频/xlsx/图片保持原文件名。 - **wiki/** — AI 维护的消化笔记。允许多级子目录。每篇文章 md 格式，命名自由（描述性即可）。 - **assets/** — 配图资源。 ## 两个特殊文件
suitedacesCLAUDE.mdsuitedaces/dorabot/CLAUDE.md
type, fill, fill_form, select, press, hover, upload_file, scroll, cookies, evaluate, console/network inspection, pdf, etc. ### Skills - `src/skills/loader.ts` — load from dirs, eligibility checks, prompt matching - `skills/` — 9 built-in skills
Junkz3CLAUDE.mdJunkz3/wrench-board/CLAUDE.md
reconcile) routes/ FastAPI routers (generate, repairs, packs, progress, documents, schematic, measurements, board_delta) schematic/ PDF schematic → page vision → merge → ElectricalGraph, plus simulator + hypothesize (deterministic engines) bench_generator/ Auto-generates simulator
versyxdigitalCLAUDE.mdversyxdigital/mkeditor/CLAUDE.md
toolbar. Exports use the File System Access API or `window.open` + `print()` for PDF. Mode branching lives in [index.ts](src/browser/index.ts), [EditorManager](src/browser/core/EditorManager.ts), [SettingsProvider](src/browser/core/providers/SettingsProvider.ts), [ExportSettingsProvider](src/browser/core/providers/ExportSettingsProvider.ts), and a few React components
timothywarner-orgCLAUDE.mdtimothywarner-org/claude-code/mcp-teaching-kit-main/CLAUDE.md
Corpus The `docs` dict in `mcp_server.py` ships six audience-calibrated documents (2 `.md`, 2 `.pdf`, 1 `.docx`, 1 `.txt`). All are 150-250 words with embedded structure cues
franklee16CLAUDE.mdfranklee16/academic-research-skills/CLAUDE.md
tools/` | 10 | Posters, web, LaTeX tools, case writing, diagrams | | `project-management/` | 4 | Project scaffolding, PDF tools, strategic analysis | ## Working with Skills ### Adding a New Skill 1. Create a folder inside
RLabs-IncCLAUDE.mdRLabs-Inc/gemini-mcp/CLAUDE.md
summarization 12. **Document Analysis** (`document.ts`): - `gemini-analyze-document`: Analyze PDFs, DOCX, spreadsheets - `gemini-summarize-pdf`: Quick PDF summarization - `gemini-extract-tables`: Extract tables from documents 13. **URL Context** (`url-context.ts`): - `gemini
kklimukCLAUDE.mdkklimuk/docx-cli/CLAUDE.md
render` shells out to Word (macOS/Windows) or LibreOffice (cross-platform) to produce a PDF, then rasterizes in-process via the bundled `@hyzyla/pdfium` WASM package (no system tools needed
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
