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/llmpolska/llm-design-engine/agents-mdgit clone --depth 1 https://github.com/llmpolska/llm-design-engineWrote 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/llmpolska/llm-design-engine/agents-md)<a href="https://agentmods.dev/instructions/llmpolska/llm-design-engine/agents-md"><img src="https://agentmods.dev/badge/instructions/llmpolska/llm-design-engine/agents-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 | $0.00720 | $0.00720 |
| Opus 5 | $0.00360 | $0.00360 |
| Sonnet 5 | $0.00144 | $0.00144 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
llm-design-engine AGENTS.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 5d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
These instructions apply to coding agents working in llm-design-engine.
Product contract
LLM Design Engine is a creative director and design compiler for coding agents. The primary promise is Design before code. Do not turn it into a theme picker, template gallery, component-library wrapper, or generic AI SaaS landing page.
The pipeline is:
brief → domain interpretation → metaphor → narrative → composition
→ design AST → deterministic preview → agent instructions
Repository rules
- Use Node.js 22+ and pnpm 10+.
- Use strict TypeScript and ESM.
- Use workspace package boundaries; import contracts from
@llm-design-engine/coreand@llm-design-engine/design-formatinstead of duplicating types. - Keep providers behind interfaces. Mock/disabled modes must work without credentials.
- Use Vue 3/Vite for apps. Tailwind is an implementation utility, not a source of design decisions.
- Do not add shadcn, DaisyUI, Bootstrap, Material UI, Ant Design, or premade landing templates.
- Do not imitate named companies, living designers, or existing websites.
- Avoid purple gradients, glassmorphism, floating card farms, generic device mockups, meaningless blobs, and centered-copy defaults unless the approved direction explains them.
- No placeholder lorem ipsum, fake API success, unfinished TODO/FIXME comments, or hidden feature claims.
- Use the LLM Design Engine brand tokens: warm paper, kiln black, oxidized coral, mineral teal, registration marks, mono annotations.
Test-first workflow
Write a focused behavioral test first. Run it and confirm the expected failure. Implement the smallest change. Run the focused test, then the affected package typecheck/build. Do not rely on a snapshot of source text as the only assertion.
pnpm --filter @llm-design-engine/<package> test
pnpm --filter @llm-design-engine/<package> typecheck
pnpm --filter @llm-design-engine/<package> build
Before completion run ./node_modules/.bin/eslint ., pnpm typecheck, pnpm test, and the app builds. The harness wrapper around pnpm lint can emit JSON help text; direct ESLint output is authoritative.
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.
- 5d ago First seen · 52 lines · 720 tokens per session scan A 492d92396f1c
llm-design-engine AGENTS.md is an instructions file published in the GitHub repository llmpolska/llm-design-engine (5 stars, last pushed 25d ago), licensed MIT. It adds 720 tokens to every session, about $0.0036 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 2026-08-31.
Other instructions, from other repositories
Orkas-VideoStudio AGENTS.md
Instructions for Orkas-AI/Orkas-VideoStudio, covering working with orkasvideostudio (for coding agents), skills and capabilities.
OrchestrUI AGENTS.md
AGENTS.md instructions for ECD5A/OrchestrUI, covering orchestrui repository instructions, mission, source of truth, legal guardrails and routing principles.
OrchestrUI CLAUDE.md
Claude Code instructions for ECD5A/OrchestrUI: Read AGENTS.md first. For UI work, use the router, orchestrator and quality-audit skills under .agents/skills. Prefer current official upstream sources over memory. Never vendor React Bits or paid/Pro content, and preserve the host project's design language unless…
harvey-design-system-mcp AGENTS.md
AGENTS.md instructions for thenpceo/harvey-design-system-mcp: Run the local server yourself and open the preview in the browser available to this environment. Do not give the user server-start instructions when you can run it.
daisyui code_generation_with_git.instructions.md
Instructions for saadeghi/daisyui, covering activation, initial setup, development loop, error handling and session end.
starwind-ui AGENTS.md
Instructions for starwind-ui/starwind-ui, covering agents.md, project shape, useful commands and contribution conventions.