F
FolderAI
Работает · 110 файлов · 7 фич за эту неделю

Массовый редактор файлов
для AI.

Одно предложение превращается в готовую к запуску папку agent folders. Потом редактируй десятки файлов за раз.

Начать бесплатно →Смотреть пайплайн ↓

Бесплатный план · вход по magic link · работает с Claude, GPT, Gemini

$79 · Ch1 freeновое

Prompting Sentinel

The world's strongest prompting simulator. Specs, dials, defenses, ship.

$99 · Ch1 free

Automation AI Engineer

Ralph-loop, TDD prompts, parallel worktrees, self-healing pipelines.

folder/sales-agent-clothing-brand
├ core/
│  ├ identity.md      ← brand voice + tone
│  ├ rules.md         ← what we will & won't do
│  └ pipeline.md      ← end-to-end flow
├ agents/
│  ├ qualifier.md     ← scores incoming leads
│  ├ writer.md        ← drafts the reply
│  └ reviewer.md      ← checks for off-brand language
├ workflows/
│  └ inbox_triage.md  ← step-by-step on new DM
└ tools/
   ├ catalog_lookup.md
   └ size_chart.md

Проблемы, которые мы решаем

Так больше не надо.

F1Решено

Vague intake

3-question quizzes can't capture a serious system.

Adaptive 7-step onboarding for pro & business tiers.

F2Решено

AI guesses too much

Generators spit out files without ever asking what you mean.

Interactive CLARIFY dialog — AI grills you with targeted questions before generating.

F2.5Решено

Endless typing on clarify

Answering 5 questions over chat is friction.

AI proposes complete answers — one click to fill, edit if needed.

F3Решено

Output locked in the app

You can see your folder but you can't take it anywhere.

One-click ZIP — production-ready folder, plug into Claude / GPT / Gemini.

F4 + F5Решено

Can't iterate, can't see what it does

Generated and walked away — now what?

Refine via chat (AI updates files in place) + capability map of every role.

🔥Решено

Generated ≠ correct

Beautiful files. Are they consistent? Complete? Aligned with your goal?

Grill folder mode — AI critiques itself, surfaces gaps, contradictions, dead refs.

Под капотом

8 этапов, полностью в прямом эфире.

Наблюдай за каждым этапом в реальном времени. Останови, разветви или уточни на любом шаге.

  1. ЭТАП 1
    QUIZ
    →
  2. ЭТАП 2
    REQUEST
    →
  3. ЭТАП 3
    CLARIFY
    →
  4. ЭТАП 4
    ARCHITECT
    →
  5. ЭТАП 5
    WRITER
    →
  6. ЭТАП 6
    OPTIMIZER
    →
  7. ЭТАП 7
    CONFLICT
    →
  8. ЭТАП 8
    EXECUTION

От mihalkevich.com · сделано с FolderAI

Симуляторы — попробуй перед тем, как строить.

Бесплатные площадки для трафика. Платные треки для навыков. B2B-спецификации для команд. Каждая карточка работает на одном движке FolderAI.

3 бесплатных · без регистрации13 платных трека · от $94 B2B-спека · под запрос

Free simulators(3)

Try without signup. Built to leak into Telegram & X.

Freeбез регистрации

Promt Enhancer Lite

Paste any prompt — get the upgraded version with 5 best-practice rules applied.

  • Specificity, context, output format, constraints, examples — applied automatically
  • Side-by-side before/after view
  • Works with Claude, GPT, Gemini system prompts
Try it →mihalkevich.com
Freeбез регистрации

Spec Linter Lite

Drop a spec → see exactly what an AI would do with it. 3-layer validator: format, autotest, rubric.

  • Format check: schema, PII, source attribution
  • Auto-test: minimal prompt run on a target model
  • Rubric grade: clarity, modularity, model-fit
Lint a spec →mihalkevich.com
Freeбез регистрации

Folder Sniff

Upload an existing AI folder — get the 3 most critical issues, ranked, with fix suggestions.

  • Detects dead refs, duplicate roles, missing safety rails
  • Quotes the exact lines that fail
  • One-click handoff to the FolderAI editor
Sniff a folder →mihalkevich.com

Education tracks(13)

Self-paced courses with auto-graded exercises.

$19$19

Agent for Human

7-day course: hire your AI. From SOUL.md to a real working colleague — with daily exercises auto-graded by FolderAI.

  • Day 1–2: SOUL.md & IDENTITY.md — voice, role, scope
  • Day 3–4: 3-layer memory & PARA knowledge graph
  • Day 5: Tools & approval queue
Try Lite free →mihalkevich.com
$39$39

Specification for AI

Write specs that an AI can't misread. PRD-as-tests, acceptance criteria, deterministic checks.

  • PRD = markdown checklist + failing tests + sample data
  • Acceptance criteria that survive non-deterministic agents
  • Live evaluator with weighted rubrics
Try Lite free →mihalkevich.com
$49$49

I'm Agent Creator

Build agent folders end-to-end: pipeline, sub-agents, memory layers, conflict detection.

  • Design 8-stage pipelines (QUIZ → EXECUTION)
  • Sub-agent patterns: when to spawn, when to merge
  • 3-layer memory in every folder you ship
Try Lite free →mihalkevich.com
$99$99

Automation AI Engineer

Ship like a team of one. Ralph-loop, TDD prompts, parallel worktrees, self-healing pipelines.

  • Ralph-loop runner: many short sprints instead of one long marathon
  • TDD prompts — failing tests first, then implementation
  • Parallel agents in tmux + heartbeat health checks
Start Ch1 free →mihalkevich.com
$79$79

Prompting Sentinel

The world's strongest prompting simulator. Specs, context, dials, examples, schemas, lies, adversaries, ship.

  • Ch1: Specs over vibes — write the eval before the prompt
  • Ch2-3: Context engineering + thinking-budget dials
  • Ch4-5: Few-shot mastery + output schemas
Start Ch1 free →mihalkevich.com
$89$89

AI Ops Analyst

Make the numbers honest. Interrogate metrics, audit triage agents, write reconciliation tests, brief AI that surfaces bad news instead of burying it.

  • Ch1: The dashboard is lying — catch a metric that lies before the board does
  • Interrogate a definition, not a direction — kill vanity metrics
  • Audit a triage agent that auto-closed the real incident as noise
Start Ch1 free →mihalkevich.com
$9 / $29 / $79$9 / $29 / $79

AI Content Creator — Start · Medium · Pro

Three tiers of a content factory: from a single funnel to a full A/B-tested production line.

  • Start ($9): one funnel, 7 ready post templates
  • Medium ($29): + content calendar, persona library, brand voice
  • Pro ($79): + multi-model A/B test, evaluator rubrics, weekly auto-refresh
Try Lite free →mihalkevich.com
$99$99

Vibe Coding Engineer

Ship a real product with Claude Code + Codex from idea to a live URL — Vercel, Railway, and a working CI.

  • Drive Claude Code & OpenAI Codex like a senior pair, not a slot machine
  • From empty repo to deployed app on Vercel / Railway in one sitting
  • Wire a CI gate so the agent's work fails loudly before it ships
Join the waitlist →mihalkevich.com
$89$89

Claude Code Power User

Master the workspace: subagents, MCP servers, hooks, slash commands, and skills that turn one terminal into a team.

  • Design subagent fleets that fan out and verify each other's work
  • Build and wire MCP servers to your own tools and data
  • Automate with hooks, custom slash commands, and reusable skills
Join the waitlist →mihalkevich.com
$129$129

AI Agent Architect

Design agents that remember, plan, and use tools — memory, skills, orchestration, and the eval loop that keeps them honest.

  • Architect memory, skills, and tool-use for long-running agents
  • Choose single-agent vs multi-agent without cargo-culting frameworks
  • Build the eval harness that catches regressions before users do
Join the waitlist →mihalkevich.com
$119$119

AI Solopreneur

Run a one-person company on agents: validate an offer, build it, price it, and find your first ten paying customers.

  • Validate demand before you build with cheap, honest experiments
  • Stand up the product, billing, and support loop solo with AI
  • Price on value and model the unit economics that keep you alive
Join the waitlist →mihalkevich.com
$89$89

MCP & Tools Developer

Build the connective tissue: production MCP servers and tool integrations that let any agent touch your systems safely.

  • Build MCP servers from scratch — resources, tools, and prompts
  • Connect agents to databases, APIs, and internal services safely
  • Design tool schemas an agent can't misuse or misread
Join the waitlist →mihalkevich.com
$99$99

LLM Evals Engineer

Stop shipping on vibes: build eval harnesses, LLM-as-judge rubrics, and regression sets that make model quality measurable.

  • Build datasets and rubrics that actually predict real quality
  • Run LLM-as-judge without fooling yourself on the scores
  • Catch regressions across prompt and model changes automatically
Join the waitlist →mihalkevich.com

Specs for teams & companies(4)

Quiz-driven specifications with acceptance test suites.

B2B

AI Work Spec

Adaptive quiz → tailored AI-Work specification, with five live test runs proving it works.

  • Pro/Business adaptive quiz (extra steps when complexity rises)
  • PDF + Markdown deliverable, plug-and-play into Claude / GPT
  • Acceptance suite included: pass/fail before you commit
Get a spec →mihalkevich.com
B2B

AI Big Data Spec

From raw signals to AI-ready pipelines. Spec covers ingestion, taxonomy, validator gates, decay.

  • Hot/Warm/Cold fact decay — never delete, supersede
  • PARA-style knowledge graph (Projects/Areas/Resources/Archives)
  • Nightly extraction cron + semantic search backend
Get a spec →mihalkevich.com
B2B

AI for Life Spec

Personal AI architecture: identity, memory, schedule, safety. Designed for one user, lived in daily.

  • SOUL + IDENTITY tuned to your voice
  • Trust ladder calibrated to your risk tolerance
  • Approval queue patterns for high-stakes actions
Get a spec →mihalkevich.com
B2B · audit

Company Digitization & AI Adoption

Audit your company → 90-day AI roadmap. Where AI saves hours this quarter, and where it shouldn't be touched.

  • Full role-by-role audit (intake quiz + interviews)
  • Per-team AI specs with acceptance tests
  • 90-day rollout plan, with cost ceilings per stage
Book the audit →mihalkevich.com
Открыть весь каталог →

Три способа начать

Собери сам, освой ремесло или закажи у нас под ключ.

01

Нанять ИИ

Опиши роль — получи переносимую папку: SOUL.md, IDENTITY.md, workflows — которую запустит любая модель.

Собрать бесплатно →02

Освоить ремесло

Сюжетные симуляторы оценивают твои решения и выдают сертификат. Первая глава бесплатна.

Смотреть треки →03

Заказать в студии

Хочешь под ключ? Делаем AI-продукты от и до и встраиваем агентов в твой стек.

Открыть студию →

Стек, на котором мы строим

Современные технологии, модели и агенты — готовые к проду

Frontier- и open-модели, агентские фреймворки, данные и инфраструктура вплоть до on-prem — то, на чём мы делаем продукты и AI-агентов.

01

Модели

Frontier и open-weights — под задачу и бюджет

  • ClaudeAnthropic
  • GPTOpenAI
  • GeminiGoogle
  • GrokxAI
  • LlamaMeta
  • MistralMistral AI
  • DeepSeekDeepSeek
  • QwenAlibaba

02

Оркестрация

Агенты, инструменты, MCP и пайплайны

  • MCPModel Context Protocol
  • LangGraphLangChain
  • LangChainLangChain
  • LlamaIndexLlamaIndex
  • CrewAICrewAI
  • Pydantic AIPydantic
  • AI SDKVercel
  • OllamaOllama

03

Данные

Базы, векторный поиск и ML

  • PostgresPostgreSQL
  • SupabaseSupabase
  • RedisRedis
  • MilvusZilliz
  • ElasticsearchElastic
  • DuckDBDuckDB
  • PyTorchPyTorch Foundation
  • Hugging FaceHugging Face

04

Инфраструктура

Облако, контейнеры и закрытые контуры

  • DockerDocker
  • KubernetesCNCF
  • TerraformHashiCorp
  • Linuxon-prem · air-gapped
  • CUDANVIDIA
  • NGINXF5
  • VercelVercel
  • RailwayRailway

Товарные знаки принадлежат их владельцам. Не партнёрство — просто инструменты, с которыми мы работаем.

Одно предложение — и рабочая папка готова.

Никакой настройки. Никакого SDK. Подключи папку к своей модели и шипи.

Начать бесплатно →