Brok's ForgeAI Engineering
Docs
GitHubHome
All documentation

Introduction

  • What is Broks Forge?
  • The AI Engineering Operating System
  • Getting Started
  • The Five Layers

Core Concepts

  • Core Concepts
  • Engineering Intelligence
  • Engineering Memory
  • Knowledge
  • Why Observability Is Not Enough
  • Deterministic Engineering Reasoning

Capabilities

  • Registry
  • AI Git
  • Forge Graph
  • Execution Graph & Failure Graph
  • Evolution
  • Brok — the Engineering Partner
  • Root Cause Explorer
  • Evaluations & Metrics

Working With Broks Forge

  • The Engineering Workflow
  • Examples
  • Best Practices

Comparisons

  • Comparisons Overview
  • Broks Forge vs LangFuse
  • Broks Forge vs LangSmith
  • Broks Forge vs Promptfoo
  • Broks Forge vs Helicone
  • Broks Forge vs Weights & Biases

Developer Documentation

  • Architecture Overview
  • Data Model
  • REST API
  • Module Structure
  • Extension Points
  • Developer Setup & Build
  • Engineering Principles

Reference

  • FAQ
  • Glossary

Engineering Handbook

  • Master Architecture
  • Engineering Handbook
  • Developer Guide
  • Project Rules
  • Coding Standards
  • API Guidelines
  • Security
  • Error Handling
  • Testing Strategy
  • Performance
  • Deployment
  • Contributing
  • Roadmap
Docs/Capabilities

Registry

The Registry is one catalog of everything you engineer: every artifact, and every piece of knowledge derived from those artifacts.

An engineering catalog, not a CRUD list

The distinction matters, and it is the reason the Registry exists as its own layer.

A CRUD list shows rows in a table: name, created date, status. One list per module, so discovery is scattered — prompts here, agents there, datasets somewhere else, and the reasoning about them nowhere at all.

The Registry shows the engineering estate: what exists, what state it is in, what has evidence behind it, and what has never been measured. Artifacts and the knowledge derived from them sit side by side, because in engineering practice they are the same question.

Two scopes

Artifacts — agents, prompts, datasets, providers, models, evaluations. Everything you registered or created.

Knowledge — observations, claims, decisions, evidence and knowledge objects. Everything the platform derived from that work.

Switching between them is one control. That adjacency is the point: "which prompts do we have" and "which of our prompt decisions have evidence" are questions the same person asks in the same minute.

What it is for

  • Discovery — one place to find anything, searchable and filterable by type.
  • Triage — see at a glance what carries a verdict and what is unknown (never measured).
  • Navigation — every entry opens its engineering workspace, on the tab that matters.
  • Orientation — a new team member can understand the shape of the estate in a few minutes.

Every entry is a doorway

Opening an artifact from the Registry lands you in its workspace, which carries the four views that matter:

TabWhat it holds
OverviewThe artifact itself
IntelligenceObservations, claims, decisions, evidence, knowledge, memory
EvolutionLineage, dependents, impact, and the AI Git timeline
Type-specificVersions, runs, execution graph, root cause — depending on the artifact

Opening a knowledge object lands you on its own page: what created it, the decision and evidence behind it, and every artifact it affects.

Brok is here too

The Registry header carries an Ask Brok entry. The natural question from a catalog view is "What is the biggest engineering risk right now?" — which is exactly what it opens with.

What the Registry deliberately does not do

  • It does not duplicate the module pages. Agents still have an Agents page; the Registry is the

cross-cutting view, not a replacement.

  • It does not invent state. An artifact with no evaluations shows as unknown, not as healthy.
  • It does not rank. There is no score. Ranking by consequence is Brok's job

("What is the biggest engineering risk?"), and it explains its reasoning when it does it.

See also: Core Concepts · Forge Graph · Evolution

PreviousDeterministic Engineering ReasoningNextAI Git