How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
vobaseCLAUDE.mdvobase/vobase/CLAUDE.md
explicit status transitions (`applyTransition` from `~/runtime`), auditable mutations. - Realtime = `pg_notify` after commit; the frontend hook invalidates matching TanStack Query keys via `table` → first-element-of-`queryKey` mapping. - Extensibility = core
lktiepCLAUDE.mdlktiep/cortex-hub/CLAUDE.md
assigned tasks ## Tech Stack Monorepo: pnpm + Turborepo. TypeScript strict. Hono API. Next.js 15 frontend. SQLite + Qdrant. Docker Compose. ## Code Conventions camelCase vars/functions, PascalCase types. `@cortex/*` path aliases. No `any` without
runbear-ioCLAUDE.mdrunbear-io/beardrive/CLAUDE.md
build -o bdrive ./cmd/bdrive # build the binary (gitignored at repo root) # web frontend (internal/webapp/frontend — only needed when changing frontend/src): npm run build # rebuild the committed assets in internal/webapp/static (run from
chaitanyyaCLAUDE.mdchaitanyya/lookout/CLAUDE.md
providers (OpenAI, Claude, Google) to analyze brand mentions across the web. ### Tech Stack - **Frontend**: Next.js 15, React 19, TypeScript, Tailwind CSS, Radix UI - **Database**: PostgreSQL with Drizzle ORM - **Authentication**: Better
wotai-devCLAUDE.mdwotai-dev/woterclip/CLAUDE.md
mechanical triage/routing, default for unlabeled issues (label: none, `is_default: true`) - **Worker personas** (Backend, Frontend, etc.) – implementation, escalate to CEO ### Key conventions - **Labels are the state machine.** `agent-working
gavishapCLAUDE.mdgavishap/omnia-vault/CLAUDE.md
markdown`, `obsidian-bases`, `json-canvas`, `obsidian-cli` | | Outward-facing updates/reports | `stakeholder-update-writing` (+ `humanizer`) | | Frontend design work | `taste`, `ui-ux-pro-max` (see `/design-setup`) | Slash commands: `/setup`, `/import`, `/catchup`, `/save
erikaflowersCLAUDE.mderikaflowers/investiture/CLAUDE.md
contributor a snapshot of the stack without reading ARCHITECTURE.md in full.] | Layer | Technology | |-------|-----------| | Frontend | [framework + version] | | Styling | [approach] | | State | [library or pattern] | | Backend | [framework, serverless, or "none"] | | Deployment | [platform
utensilsCLAUDE.mdutensils/mold/CLAUDE.md
generation_profiles -- --check bash scripts/tests/ci-routing-contract.sh bash scripts/tests/candle-single-identity.sh # every candle crate on ONE fork rev # Frontend (one Bun workspace at repo root; prettier scoped to studio/, desktop override in .prettierrc
alec-c4CLAUDE.mdalec-c4/claude-rails-dev/templates/CLAUDE.md
Context > **Project:** [Project Name] > **Stack:** Rails [Version] | Ruby [Version] ## 🛠 Tech Stack - **Testing:** [RSpec / Minitest] - **Frontend:** [Hotwire / Inertia / React] - **Database:** PostgreSQL - **Auth:** [Native / Devise] ## 🤖 Active Agents This project uses the **Claude
chatmlCLAUDE.mdchatml/chatml/CLAUDE.md
uses isolated git worktrees to run parallel AI agent sessions. Polyglot architecture: Next.js frontend, Go backend, Tauri wrapper, Node.js agent runner. ## Task Management (MANDATORY) Always create tasks with `TodoWrite` before
monk1337CLAUDE.mdmonk1337/clicodelog/CLAUDE.md
approaches this, split it before adding more. This applies to Python, JS and CSS. ## Frontend `static/js/` is vanilla JS with no build step, loaded as ordered script tags. State lives
vicarious11CLAUDE.mdvicarious11/agenttop/CLAUDE.md
src/agenttop/analysis/` — LLM engine, intent classification, recommendations - `src/agenttop/web/` — FastAPI server, optimizer, graph builder - `src/agenttop/web/static/` — SPA frontend (index.html, JS, CSS) - `src/agenttop/tui/` — Terminal UI (textual-based, alternative to web) ### Data Flow 1. **Collectors
jakubsuplickiCLAUDE.mdjakubsuplicki/codument/CLAUDE.md
senior-backend`; system or architecture decisions -> `senior-architect`; UI components, state, or performance -> `senior-frontend`; visual or aesthetic polish -> `frontend-design`; animation, gesture, or motion -> `motion-craft`; reviewing a diff
AgriciDanielCLAUDE.mdAgriciDaniel/claude-music/CLAUDE.md
dashboard** - `skills/claude-music/webapp/` holds a stdlib-only Python server (`server.py`) plus a single-file frontend (`index.html`). It shells out to `music_engine.py --progress` (NDJSON progress events on stderr) and stores per-track
AutomatosAICLAUDE.mdAutomatosAI/automatos-ai/CLAUDE.md
greenfield. The platform has 109 SQLAlchemy tables, 103 API routers, 599 frontend TS files, and a Knowledge Graph mapping every code path. **Most "new" work
kbhujbalCLAUDE.mdkbhujbal/AlphaAnalyst-open-source-autonomous-equity-research-agent/CLAUDE.md
modeler/ DCF + comps (pure Python) │ │ ├── orchestrator/ wires agents into pipeline │ │ └── api/ FastAPI routes │ └── tests/ ├── frontend/ Next.js app │ ├── package.json │ ├── app/ App Router pages │ ├── components/ │ ├── lib/ api client, utils, hooks │ ├── types/ generated
mandarwagh9CLAUDE.mdmandarwagh9/embedist/CLAUDE.md
this directory contains overlapping developer notes; keep both in sync when behavior changes. ## Commands Frontend scripts (`package.json`): - `npm run dev` — Vite dev server (port 1420, required by `tauri.conf.json
benchjackCLAUDE.mdbenchjack/benchjack/CLAUDE.md
streams results to a live web UI. ## Tech Stack - **Backend**: Python 3.11+, FastAPI, uvicorn - **Frontend**: Vanilla HTML/CSS/JS (no framework), SSE for real-time streaming - **AI Backends**: Claude Code CLI, OpenAI
myrialabsCLAUDE.mdmyrialabs/clopen/CLAUDE.md
Runtime checks exist in `bin/clopen.ts` and `backend/index.ts` ### Architecture - **Backend:** Bun + Elysia with WebSocket architecture - **Frontend:** Svelte 5 with runes system ($state, $derived, $effect) - **Database:** SQLite with migration system ### Code Style
wigtnCLAUDE.mdwigtn/wigtn-plugins/CLAUDE.md
Wireframe / Handoff 5종 생성) → /implement (triage → quick-fix: 경량 단일 에이전트 | feature/greenfield: DESIGN → BUILD → [Frontend? → design-discovery → style select] → [Linear 연동? → Epic/하위 이슈 등록 → 이슈 단위 순차 BUILD | 미연동
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.
