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 rules/agentset-ai/agentset/project-structuregit clone --depth 1 https://github.com/agentset-ai/agentsetWhat 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 | $0.01215 | $0.01215 |
| Opus 5 | $0.00607 | $0.00607 |
| Sonnet 5 | $0.00243 | $0.00243 |
| Haiku 4.5 | $0.00121 | $0.00121 |
Grade A, and why
project-structure 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 — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Structure & Monorepo Guidelines
Overview
Agentset is a Turborepo monorepo containing a Next.js web app and supporting packages. This is an open-source RAG (Retrieval-Augmented Generation) platform built with modern web technologies.
Monorepo Structure
Core Applications
apps/web/- Main Next.js application with App Router
Internal Packages (Workspace Dependencies)
packages/db/- Database schema, Prisma client, and typespackages/demo/- Demo data and templatespackages/emails/- Email templates and components (using react-email and resend)packages/engine/- RAG engine with vector stores and embeddingspackages/jobs/- Background job definitions (Trigger.dev)packages/storage/- File storage utilities (S3)packages/stripe/- Payment processing utilitiespackages/tinybird/- Data warehousing and analytics (Tinybird)packages/ui/- Shared UI components (Shadcn-based)packages/utils/- Shared utility functionspackages/validation/- Shared Zod schemaspackages/webhooks/- Webhook definitions and handlerstooling/- Development and build tooling
Package Management
Use Bun with Workspace Protocol
- Use
bunas the package manager (required by packageManager field) - Reference internal packages with
workspace:*in package.json - Use catalog: references for shared external dependencies
{
"dependencies": {
"@agentset/db": "workspace:*",
"@agentset/ui": "workspace:*",
"typescript": "catalog:"
}
}
Workspace Scripts
bun dev- Start all services in watch modebun dev:web- Start web app with dependenciesbun build- Build all packagesbun lint- Run linting across workspacebun typecheck- Type check all packages
Directory Naming Conventions
Use Kebab-Case for Directories
- Directory names should use lowercase with dashes:
auth-wizard/,api-client/ - File names should use kebab-case:
auth-provider.tsx,user-settings.ts - Avoid camelCase or PascalCase for directories
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 · 190 lines · 1,215 tokens per session scan A ccd8789d4223
project-structure is a cursor rule published in the GitHub repository agentset-ai/agentset (2,079 stars, last pushed 1mo ago), licensed MIT. It adds 1,215 tokens to every session, about $0.0061 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 cursor rules, from other repositories
project-rules
智能体平台前端项目开发规则和最佳实践.
cursorrules
所有 Git 提交说明必须使用简体中文书写，禁止使用英文标题或英文正文。 专有名词、包名、API 路径等技术标识可保留英文。.
write-migration
Author an idempotent, state-detecting migration under scripts/migrations/ for a breaking llm-wiki-memory change. Use when shipping a breaking change to the engine, or when asked to write a migration / upgrade path. Replaces the retired release-runbook flow.
debug-capture-pipeline
Debug llm-wiki-memory capture/recall — flush breadcrumbs, daily-vs-compile lag, failed-distill stash recovery, embedding cache, hook wiring. Use when memory didn't capture, doesn't recall, or a hook/cron looks broken.
self-observability
Self-observability (watch llm-wiki-memory while you work; offer fixes) — This rule applies to every AI agent (Claude Code, Cursor, Codex, generic MCP clients) working Canonical file is .agents/rules/self-observability.md.
vios-sqa
VIOS SQA agent. One-click build → deploy → test → UI validation cycle. Use when asked to run regression tests, check test results, deploy for testing, or verify UI behavior.