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.
git clone --depth 1 https://github.com/Scopeo/draftnrunWrote 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/rules/scopeo/draftnrun/architecture)<a href="https://agentmods.dev/rules/scopeo/draftnrun/architecture"><img src="https://agentmods.dev/badge/rules/scopeo/draftnrun/architecture.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 | $0.02822 | $0.02822 |
| Opus 5 | $0.01411 | $0.01411 |
| Sonnet 5 | $0.00564 | $0.00564 |
| Haiku 4.5 | $0.00282 | $0.00282 |
Grade A, and why
architecture 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architecture Rules
Architecture Overview
Repo Structure
ada_backend/— API server (FastAPI). Routers inrouters/, services inservices/, repos inrepositories/.engine/— Graph execution engine.graph_runner/(DAG scheduler),components/(component implementations),field_expressions/(AST).mcp_server/— Standalone MCP server wrapping the API. Separate process, no imports fromada_backend/orengine/.data_ingestion/— Document ingestion logic.workers/— Ingestion + webhook worker processes.infra/k8s/— Kustomize-based Kubernetes manifests.tests/— Test suite.
Backend Pattern
Router → Service → Repository (SQLAlchemy + PostgreSQL). Routers define HTTP endpoints and auth dependencies. Services encapsulate business logic. Repositories handle DB CRUD.
Auth Split
Supabase owns users, organizations, memberships, roles. Backend owns projects, graphs, runs, components, and everything else. Backend validates Supabase JWTs via supabase.auth.get_user(). Organization roles are checked via Supabase Edge Functions (check-org-access, check-super-admin).
API Keys
SHA-256 hashed (salted with BACKEND_SECRET_KEY), prefix taylor_, scoped to project or organization.
Graph Execution
DAG of ComponentInstances connected via FieldExpressions (JSON AST). See ada_backend/docs/engine.md.
Generated run-local files are passed by filename, not by copying bytes through graph state. Python Code Runner saves
supported generated files under get_output_dir() and exposes artifacts.files; wire that list to
AIAgent.input_filepaths when the next AI Agent should attach those files directly to the LLM request. The AI Agent
must resolve filenames relative to the run output directory and require a file-capable model.
Environments
draft / production bound to specific GraphRunner versions via ProjectEnvironmentBinding. Each Run records its graph_runner_id (set when the worker transitions the run to RUNNING).
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 First seen · 103 lines · 2,822 tokens per session scan A 5588ee9fb08b
architecture is a cursor rule published in the GitHub repository Scopeo/draftnrun (30 stars, last pushed 4d ago), licensed Apache-2.0. It adds 2,822 tokens to every session, about $0.0141 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 2026-09-04.
Other cursor rules, from other repositories
sql-queries-not-asts
Name built-and-lowered SQL units "queries", not "asts".
supabase
Do not use the Supabase Client, instead use the Jawn Client and put all new endpoints in controllers. Don't use RLS policies, instead enable RLS for new tables and revoke privledges. Example.
java-springboot-jpa-cursorrules-prompt-file
Cursor rules for Java development with Springboot and JPA integration.
fastapi-database-interaction
A set of rules for connecting FastAPI, a Python framework for building web APIs, to databases using asynchronous libraries such as asyncpg or aiomysql and SQLAlchemy 2.0.
go-service-errors
Go service error translation, nullable TIMESTAMPTZ scans, history, and side effects.
cursorrules
You are assisting the Orbital engineering team — 5 backend engineers building a multi-tenant SaaS for logistics operations. TypeScript monorepo using NestJS, PostgreSQL (drizzle-orm), and BullMQ. All engineers use shared conventions.