Black Box Architecture Agents

AI agents for enhancing software architecture using black-box design principles.

Black Box Architecture Agents

AI agents for maintaining and improving software architecture using black-box design principles — inspired by Eskil Steenberg's lecture and rooted in information hiding (Parnas, 1972).

Five coordinated specialists — analyze, plan, implement, debug — governed by a shared contract that enforces scope discipline, evidence minimums, approval gates, and a self-improvement loop.

Quick Install

git clone https://github.com/gl0bal01/black-box-architecture.git

# Project-specific
mkdir -p .claude/agents
cp -r black-box-architecture/agents/* .claude/agents/

# Or personal (all projects)
cp -r black-box-architecture/agents/* ~/.claude/agents/

Agents

AgentRoleModelWrites Code
arch-orchestratorEntry point. Classifies, delegates, synthesizes.OpusNo
arch-analyzerMaps module boundaries, coupling, violations.SonnetNo
arch-plannerDesigns modules, interfaces, migration roadmaps.OpusNo
arch-implementerExecutes approved plans with commit discipline.SonnetYes (requires approval)
arch-debuggerIsolates bugs to contract boundaries, fixes autonomously.SonnetYes (approval gates apply)

All agents follow agents/BLACKBOX_CONTRACT.md — 13 rules covering scope discipline, evidence requirements, approval gates, commit checkpoints, and self-improvement.

Usage

"Ask arch-orchestrator to analyze UserService and propose black-box boundaries"
"Ask arch-analyzer to find coupling issues in src/services/"
"Ask arch-planner to design a payment processing module"
"Ask arch-implementer to execute the approved refactoring plan"
"Ask arch-debugger to isolate and fix the checkout bug"

Workflows

WorkflowAgentsApproval Gate
Analysis onlyAnalyzer
Planning onlyPlanner
Full refactoringAnalyzer → Planner → ImplementerAfter plan, before implementation
Debug & fixDebugger → (Implementer)Before non-trivial fixes
Complete transformationAll agents in sequenceAfter plan + between phases

Documentation

License

MIT