🧭 INEMA Command Language
An operational language for 100 commands that controls how AI models and agents research, verify, analyze, critique, decide, plan, execute, and communicate.
New here? One DSL (Domain-Specific Language) is a small language built for a
specific domain—here, the domain is direct AI with precision. You type
/comando, and the command
carries a complete work protocol.
Illustration of the typical flow: you provide the goal, and the commands execute specialized protocols in sequence.
The 6 tracks
Each command family has its own learning path. Start with Fundamentals if this is your first time using the language; if you already use commands day to day, go straight to the path that addresses your bottleneck.
🧭 Language Fundamentals
Syntax, the 10 rules, chains and presets—and the family of truth and verification that supports everything else.
🛰️ Research & Analysis
Conduct structured research, detect weak signals, map ecosystems, and turn raw information into a coherent view.
🥊 Critique & Decision
Personal red team, premortem, biases—and the decision-making arsenal: ranking, scoring, matrix, Pareto, and bets.
📐 Planning & Execution
Blueprint, roadmap, architecture, and SOPs—then execute, prioritize, monitor, and review against criteria.
💼 Business & applied AI
Market, offer, pricing, and moat—and building AI products and agents with eval and guardrails.
🗣️ Communication & Mastery
Executive brief, progressive instruction, and audience adaptation—plus chains, presets, and the v0.2 roadmap.
How studying works here
Read one topic at a time
Each topic teaches a command or concept with an example ready to copy and run in your favorite agent.
Highlight important passages by selecting the text, and record a question with one click where it came up.
Mark as read and track your progress
Progress appears at three levels—module, track, and entire course—and is saved in your browser. No account required.
Return through My Journey
The dashboard brings together where you left off, your open questions, and your notes—with export/import in .json to take your study to another computer.
Recipes to use today
Official compositions from the language itself. Paste the line into your agent, replace what’s marked, and compare the response with the expected output pattern.
🎯 Reliable decisions
Objective: turn a vague choice into a recommendation with evidence, explicit gaps, and adversarial critique.
🏗️ Agent architecture
Objective: specify a complete agent before writing the first line of the prompt.
/agent /tools /memory /guardrails /eval /integrate for
How to verify: you should receive an objective, tools with usage criteria, a memory policy, operational limits, test cases, and an integration plan.
🛡️ Truthfulness audit
Objective: check a text full of claims before publishing it or making a decision based on it.
/factcheck /confidence /provenance no texto:
How to verify: each claim classified (confirmed, imprecise, false, inconclusive, opinion), overall confidence level, and source of each critical data point.