Notes from the Fabriq team
Architecture decisions, CDR engineering, and how we keep AI agents honest.
Subscribe via RSSA Feature's Journey: From Idea to Ship in Fabriq
Follow a fictional 'Order Search' feature through Fabriq's five-stage pipeline — from kickoff to deployment — and see how the workspace shell makes every phase legible, reviewable, and reversible.
AI Scanner, Engine Validator: Why We Don't Ask LLMs to Count Lines
Pure static analysis is brittle. Pure LLM analysis hallucinates. Fabriq's third path draws a clean line: Tree-sitter and ASTs answer 'where is X called?' deterministically; LLMs answer 'why is X called?' with fuzzy reasoning. The orchestrator ties them together with evidence chains that are auditable by construction.
How to Contribute to Fabriq: Starting with a Blog Post
The best way to contribute to a project is the one that matches your skills and interests. Not everyone starts with a pull request. This guide maps the full contribution landscape — from writing docs to building plugins to fixing bugs — with concrete entry points for each path.
Open Source AI Tooling: Fabriq's Strategy and Philosophy
Fabriq is MIT-licensed and community-built. This post explains our open source philosophy: why MIT, how we think about commercialization, what 'open core' means to us, and why we believe AI development tools must be transparent to be trustworthy.
Three Roles, One Tool: How Tech Leads, Engineers, and PMs Use Fabriq
Different roles have different problems with AI-assisted development. Tech leads worry about governance. Engineers want flow without context switching. PMs need traceability from intent to outcome. Fabriq serves all three — through the same workspace, from different angles.
Cursor vs Fabriq: Why Agent-Agnostic Architecture Wins
Most AI coding tools lock you into one agent. Fabriq's agent-agnostic architecture works with OpenCode, Claude Code, or any ACP-compatible backend — and the protocol is more important than any single model vendor.
Local-First vs Cloud-First: Why AI Dev Tools Should Run on Your Machine
The AI coding industry trends toward cloud-hosted agents. Fabriq goes the other way: local-first, offline-capable, putting data sovereignty and team control above convenience. Here's the reasoning behind that bet.
The Naming of Fabriq: From dapei to Workshop
A name carries a philosophy. Fabriq, dapei, fabric — the thread that connects them traces the evolution of an idea about how humans and AI should build software together.
Software Engineering's Last Black Box
Version control solved 'who changed what'. Nobody solved 'why is this code here'. Fabriq's evidence-first approach treats every code change as a claim that must be sourced, validated, and attributable — making AI-generated code auditable by default.
Typed IPC: Why Your Electron App Needs a Type-Safe Customs Office
We killed the generic 'run any capability' channel from the renderer after a threat model review. Every IPC call in Fabriq now goes through a typed, Zod-validated handler. Here's the threat model, the migration, and how it makes the app auditable by default.
Why AI Coding Agents Need a Control Panel, Not Another Chat Box
AI agents edit files, run tools, and ship changes in seconds — but you're the one accountable for what lands in the repo. Most tools give you a chat box. Fabriq gives you a workshop where every agent action is legible, reviewable, and reversible.
Why We Chose Electron (and We're Not Sorry)
Electron gets a bad rap. For a v0.1 desktop shell that must integrate deeply with local processes, Node.js native modules, and external LLM backends — the alternatives were worse. Here's the trade-off analysis that led to our decision.
CDR from dapei-skill to Fabriq Desktop: Four Hard-Earned Moves
CDR keeps code knowledge alive by 'AI as Scanner, Engine as Validator.' In dapei-skill we shipped it as a headless engine powering a static portal. In Fabriq Desktop we kept the philosophy but rewired the architecture around symbol handles, an embedded engine, and a strict workspace/feature boundary.
When to Open a Feature — and When Not To
A Feature is the only unit of work that can be shipped end-to-end. This post spells out the rule, the anti-patterns, and the artifact trail every Feature should leave behind.
Fabriq Product Blueprint: From Workspace OS to Feature Delivery Loop
Fabriq is building a local-first execution system for product engineering: stable workspace assets, isolated feature execution, confirmation-driven closure, and auditable knowledge promotion.
Fabriq Roadmap Update (2026-06): What Shipped, What Is In Flight
A transparent progress snapshot of Fabriq roadmap: shipped foundations, active workstreams, and the next milestones we are prioritizing.
Workspace + Feature: Why Two Dimensions Beat One
Most tools collapse everything into a single timeline. We split the world into a workspace dimension and a feature dimension — and the boundary is what makes confirmation-driven automation possible.
AI Scanner, Engine Validator: Stop Asking LLMs to Count Lines
Tree-sitter and ASTs answer 'where is X called?' deterministically. LLMs answer 'why is X called?' with fuzzy reasoning. Fabriq draws the line in the right place — and the result is audit-ready cognitive artifacts.
Why We Killed the Generic capability.run Channel from the Renderer
ADR-0021: read-only capabilities are the only ones the renderer can invoke directly. Everything else goes through a typed IPC handler with a Zod-validated payload. Here's the threat model and the migration path.