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.
npx agentmods add instructions/cliftonc/drizzle-cube/claude-mdgit clone --depth 1 https://github.com/cliftonc/drizzle-cubeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/cliftonc/drizzle-cube/claude-md)<a href="https://agentmods.dev/instructions/cliftonc/drizzle-cube/claude-md"><img src="https://agentmods.dev/badge/instructions/cliftonc/drizzle-cube/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02055 | $0.02055 |
| Opus 5 | $0.01027 | $0.01027 |
| Sonnet 5 | $0.00411 | $0.00411 |
| Haiku 4.5 | $0.00205 | $0.00205 |
Grade A, and why
drizzle-cube 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 2d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Drizzle ORM-first semantic layer with Cube.js compatibility. Type-safe analytics and dashboards with SQL injection protection.
Project Structure
src/server/ # Core semantic layer — compiler, executors, query planning → src/server/CLAUDE.md
src/client/ # React analytics dashboard components → src/client/CLAUDE.md
src/adapters/ # Framework adapters (Express, Fastify, Hono, Next.js) → src/adapters/CLAUDE.md
src/i18n/ # Internationalization runtime and locale files → src/i18n/CLAUDE.md
src/cli/ # CLI tool — `npx drizzle-cube charts init|list` → src/cli/CLAUDE.md
src/shared/ # Shared utilities (date-range parsing)
tests/ # Multi-database testing infrastructure → tests/CLAUDE.md
dev/ # Development environment: example server, Docker Compose, migrations, seed
.claude/ # Agent skills
Agent Documentation Context
Before planning non-trivial changes, review the LLM-optimized documentation index at https://www.drizzle-cube.dev/llms.txt for current public docs, examples, and API context. Use it as supplementary context alongside this repository's CLAUDE.md files and source code; repository files remain the source of truth for implementation details.
Database Support
PostgreSQL · MySQL · SQLite · DuckDB · Snowflake · Databend · SingleStore
Each engine has a dedicated executor in src/server/executors/. Auto-detection resolves the engine from the Drizzle instance.
Essential Commands
| Category | Command | Description |
|---|---|---|
| Dev | npm run dev |
Start dev servers (server + client) |
| Dev | npm run dev:server |
Watch server (tsx) |
| Dev | npm run dev:client |
Watch client (Vite) |
| Dev | npm run dev:setup |
Start Docker DBs, run migrations + seed |
| Build | npm run build |
Build all (server, client, adapters, CLI) |
| Build | npm run build:server |
Build server only |
| Build | npm run build:client |
Build client only |
| Build | npm run build:adapters |
Build adapters only |
| Build | npm run build:cli |
Build CLI only |
| Test | npm test |
Run all tests (Vitest) — needs Docker, see below |
| Test | npm run test:watch |
Tests in watch mode |
| Test | npm run test:postgres |
Run server tests (PostgreSQL, default engine) — needs Docker |
| Test | npm run test:mysql |
Run server tests against MySQL — needs Docker |
| Test | npm run test:sqlite |
Run server tests against SQLite — no Docker |
| Test | npm run test:cli |
Run CLI / generator tests — no Docker |
| Test | npm run test:client |
Run client component tests — no Docker |
| Test | npm run test:e2e |
Run Playwright end-to-end tests |
| Test | npm run test:setup |
Start test containers (docker-compose up) |
| Test | npm run test:teardown |
Stop test containers (docker-compose down) |
| Quality | npm run typecheck |
TypeScript type checking |
| Quality | npm run lint |
ESLint |
| Quality | npm run lint:fix |
ESLint with auto-fix |
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.
- 2d ago Changed · +1 lines · +86 tokens per session ea70e3feac50
- 5d ago First seen · 124 lines · 1,969 tokens per session scan A dde2093fe393
drizzle-cube CLAUDE.md is an instructions file published in the GitHub repository cliftonc/drizzle-cube (60 stars, last pushed today), licensed MIT. It adds 2,055 tokens to every session, about $0.0103 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
data-table-filters CLAUDE.md
Claude Code instructions for openstatusHQ/data-table-filters, covering claude.md, project overview, monorepo structure (pnpm + turborepo), key commands and database.
better-drizzle AGENTS.md
AGENTS.md instructions for almeidazs/better-drizzle, covering better-drizzle repository – agent field notes, architecture, design intent, api surface and performance rules.
better-drizzle CLAUDE.md
Claude Code instructions for almeidazs/better-drizzle: Use AGENTS.md as the repository-wide source of truth.
better-drizzle GEMINI.md
Gemini CLI instructions for almeidazs/better-drizzle: Use AGENTS.md as the repository-wide source of truth.
build-applications-w-copilot-agent-mode octofit_tracker_setup_project.instructions.md
Instructions for skills/build-applications-w-copilot-agent-mode, covering octofit tracker multi-tier application setup guidelines, application goals, command execution rules, forwarded ports and project structure.
counterpoise-ledger CLAUDE.md
Instructions for jeffjjohnston/counterpoise-ledger, covering claude.md, project overview, development commands, essential commands and testing.