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/n-shadloo/frontend-production-engineer/gemini-mdgit clone --depth 1 https://github.com/n-shadloo/frontend-production-engineerWrote 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/n-shadloo/frontend-production-engineer/gemini-md)<a href="https://agentmods.dev/instructions/n-shadloo/frontend-production-engineer/gemini-md"><img src="https://agentmods.dev/badge/instructions/n-shadloo/frontend-production-engineer/gemini-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 | $0.01827 | $0.01827 |
| Opus 5 | $0.00914 | $0.00914 |
| Sonnet 5 | $0.00365 | $0.00365 |
| Haiku 4.5 | $0.00183 | $0.00183 |
Grade A, and why
frontend-production-engineer GEMINI.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 yesterday.
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.
This is a copy
92% identical to frontend-production-engineer AGENTS.md — 100 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gemini CLI context: frontend-production-engineer
This repo's frontend instructions live in SKILL.md and references/. Load
SKILL.md first, then the relevant references/*.md file. See AGENTS.md
for the full description. Do not duplicate the content here — read the source
files.
Coverage is production Next.js and TypeScript against a Django / Django REST Framework backend. The subjects are routing and rendering, the React component tree, and the typed contract with the backend. They also include the non-functional guarantees a shipped frontend owes its users. The Django backend itself is out of scope.
Twenty-four domains are integrated across four tiers. Tier 1 is the foundations: App Router architecture, the type system, React component architecture, and project structure and tooling. Tier 2 is the backend contract and the state built on it: DRF contract and OpenAPI codegen, data fetching, auth, and realtime. Tier 3 is interface craft: design system, accessibility, forms, tables, media, motion, and UX writing. Tier 4 is the non-functional guarantees: Web Vitals, frontend security, SEO, internationalization and RTL, testing, observability, build and deploy, and analytics and consent. A Tier 0 operating doctrine sits under all four, and it is always in effect.
At 1.24.1 the integrated material in references/ is the App Router
foundation, the type system, and the React component tree. It also holds
the project structure, the DRF contract, and the client cache and state. It
holds the session with the gates over it, and the push transport with the
events on it. It holds the design system and accessibility — the tokens and the
layout, the element and its name, the keyboard, and the measurable criteria.
It holds forms, the dense data surface, media, and motion. That part is the schema and the bind, and the table with the server that drives it. It also holds the file that leaves for a user, and the animation over all of them.
It holds the copy — the words on a control, the message that a failure produces, and the string as data behind a key.
It holds speed as a measured property — the thresholds, the budget, and the gate over them. It also holds the bytes of JavaScript, with the third-party script beside them. The last part is the largest paint, the layout shift, and the answer to a tap.
It holds the browser-side threat model, and that part holds a veto. It is the sink that turns data into code, and the sanitiser in front of it. It also holds the policy and the header set that the response carries. The endpoint that the network reaches, and the destination that the server reaches back, are in it too. The last of it is what must never cross to the browser, and the code that arrives there that nobody on the team wrote.
One part is what a machine reads. It is the title, the canonical link, and the card that a share preview renders. It also holds the claim that a page makes in machine-readable form, and the escape that the grammar of that block needs.
The last part is the crawl and the index. It holds the robots rules, and the sitemap that the DRF data produces. It also holds the status code that a missing record returns, and the redirect that a rename ships.
Another part is the second language. It holds the route that carries a locale, and the file that holds the catalog behind it. It also holds the date, the number, and the calendar that each locale writes in its own form. The last of it is the direction of a surface, and the script that a font must carry.
The next part is the proof. It holds the level that each test belongs to, and the component test that carries most of the value. It also holds the answer that a mock gives to a request, and the journey that a real browser runs. The last of it is the gate that a change passes before a merge.
Another part is what happens when something breaks. It holds the report that a failure sends, and the personal value that the report must never carry. It also holds the identifier that joins one screen to one Django log line, and the trace that crosses to the backend. The last of it is the application under an outage. That part is the gate over a dead backend, the offline state, and the probe that answers for the chain.
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.
- yesterday Changed 1d9e286ce97c
- 5d ago First seen · 141 lines · 1,827 tokens per session scan A c3390baa6b2e
frontend-production-engineer GEMINI.md is an instructions file published in the GitHub repository n-shadloo/frontend-production-engineer (2 stars, last pushed 2d ago), licensed MIT. It adds 1,827 tokens to every session, about $0.0091 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to frontend-production-engineer AGENTS.md, differing in 100 lines, and is treated as a copy.
Other instructions, from other repositories
motion-lexicon AGENTS.md
AGENTS.md instructions for Ryan-yang125/motion-lexicon, covering agent instructions, operating context, communication, product boundary and technical stack.
Kokoro-Engine CLAUDE.md
Claude Code instructions for chyinan/Kokoro-Engine, covering claude.md, git 规范, 项目概述, 常用命令 and 安装依赖.
retry-axios GEMINI.md
Gemini CLI instructions for JustinBeckwith/retry-axios, covering project: retry-axios, project overview, building and running, dependencies and commands.
openmcp-client CLAUDE.md
Instructions for LSTM-Kirigaya/openmcp-client, covering claude.md, development commands, setup and installation, development and service development.
bedrock CLAUDE.md
Claude Code instructions for christopher-buss/bedrock, covering bedrock project context, what is bedrock?, architecture, architecture quick reference and testing requirements (non-negotiable).
free-ai-gateway AGENTS.md
Instructions for zaber-dev/free-ai-gateway, covering 🤖 free-ai gateway - agentic development guidelines, 🏛️ monorepo architecture & package boundaries, 🛑 strict architectural rules for agents, 💻 essential developer commands and build all packages across monorepo.