Master Spec
Core thesis, product definition, seven platform layers, and product principles.
Read spec →Documentation
Public specs and architecture docs.
These documents define the foundation of Intent OS. They are public so builders, partners, and pilot teams can understand the system before it ships.
Strategy & Architecture
Information Architecture
Recommended routes, homepage section order, navigation labels, and content tone.
Read spec →Product Architecture
Cloudflare-first mapping, agent roles, and runtime flow from intent to outcome.
Read spec →Data Model
Core entities and data design rules for the platform.
Read spec →Policy Engine
Policy domains, required outcomes, and safety principles.
Read spec →Roadmap
12 to 24 month build plan from website launch to autonomous pilots.
Read spec →Intent OS - Technical Specs
MVP Build Order
Phase-by-phase build sequence with done criteria, dependency map, and test strategy.
Read spec →Database Schema
Full D1 schema for MVP: tenants, workspaces, runs, payments, devices, proofs, audit.
Read spec →API Specification
Unified API contract for Control Plane, Runtime, Payments, and Device Gateway.
Read spec →Admin Dashboard Spec
Screen map, personas, authorization matrix, and UX patterns for the admin UI.
Read spec →Admin UI Routes & Screens
Detailed route definitions, filter bars, table UX, and empty state patterns.
Read spec →Run Detail Page
The most important screen: lifecycle, steps, approvals, payments, proofs, reconciliation.
Read spec →D1 SQL Migrations V1
Migration plan, file order, seed data, and execution strategy for Cloudflare D1.
Read spec →Backend Folder Structure
Monorepo layout, module boundaries, naming conventions, and testing targets.
Read spec →Setup
Cloudflare + GitHub Setup
Repo structure, Pages configuration, DNS, and security baseline.
Read guide →AI Governance
AI Governance Index
Single reading path across Level 1, Level 2, and Level 3 AI operating documents.
Read index →AI System Rules
Core AI operating rules, constraints, workflow, and output standards for this repository.
Read rules →Prompt Templates
Reusable prompt formats for analysis, planning, building, debugging, review, and docs work.
Read templates →Task Handoff Template
Structured handoff format for assigning scoped work to AI agents and collaborators.
Read template →Context Memory System
Rules for summary reuse, derived memory, and minimizing repeated context loading.
Read system →Agent Execution System
Role-based AI workflow for architect, builder, debugger, reviewer, and docs execution.
Read system →Multi-Model Router
Model selection strategy for simple, standard, complex, bulk, and large-context tasks.
Read router →Execution Engine (Level 3)
Autonomous task execution flow from classification and splitting to validation and merge.
Read engine →Realtime Agent System
Coordination rules for parallel agents, conflict control, observability, and live handoff.
Read system →Data Intelligence Layer
How AI ingests, indexes, retrieves, and reasons over repository and runtime data.
Read layer →Contact
Questions about the specs?
We welcome builders, partners, and researchers who want to dig deeper.