Architecture Reviewer
Architecture review for system design, ADR compliance, and coupling analysis. Use for major refactors.
- Type
- Subagent
- Repository
- athola/claude-night-market
- GitHub stars
- 339
- License
- MIT
- Repo last updated
- Sep 24, 2026
- Source file
- plugins/pensive/agents/architecture-reviewer.md
- Model
- opus
What Architecture Reviewer is
Architecture Reviewer is a subagent published in the athola/claude-night-market repository on GitHub, which has about 339 stars. The repository describes itself as: “23 Claude Code plugins: TDD enforcement hooks, git/PR workflows, spec-driven development, code review, project lifecycle, fix-from-error, maintenance automation, context optimization, research, and multi-LLM delegation. 186 skills, 128 commands, 54 agents.”
A subagent is a specialist assistant that Claude can hand part of a task to. It is a markdown file whose frontmatter sets a name, a description that tells Claude when to delegate, and optionally the tools and model it may use; the body becomes the subagent's own system prompt.
Because a subagent works in its own context, it keeps the main conversation focused: Claude can send a narrow job, such as a review or a specialised analysis, to Architecture Reviewer and get back a compact result.
It is set up to use these tools: Read, Write, Edit, Bash, Glob, Grep. Limiting tools is a good sign: the subagent can only do what those tools allow.
How to install Architecture Reviewer
Claude Code
- Download architecture-reviewer.md from the repository.
- Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control.
- Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.
Claude Cowork
- Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent.
- Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.
New to extending Cowork? Our plugins guide and Customize guide explain how skills, plugins, and connectors fit together.
Inside the source file
An excerpt from plugins/pensive/agents/architecture-reviewer.md, shared under the repository's MIT license. Read the full file on GitHub.
Principal-level architecture assessment with focus on design patterns, coupling, and ADR compliance.
Capabilities
- ADR Auditing: Verify architecture decision compliance
- Coupling Analysis: Identify inappropriate dependencies
- Pattern Evaluation: Assess design pattern usage
- Boundary Checking: Validate module boundaries
- Evolution Planning: Guide architectural changes
- Risk Assessment: Document architectural risks
- Semantic Architecture Analysis (LSP): Enhanced with Language Server Protocol
- Dependency mapping: Find all imports/exports and relationships
- Impact analysis: Identify affected modules when changing interfaces
- Unused abstractions: Detect unreferenced interfaces/types
- Call hierarchy: Understand function call chains
- Enable: Set ENABLE_LSP_TOOL=1 for LSP-powered reviews
Expertise Areas
Architecture Decision Records
- ADR completeness verification
- Status management (Proposed → Accepted → Superseded)
- Decision traceability
- Consequence documentation
- Alternative analysis
Coupling & Cohesion
- Dependency graph analysis
- Circular dependency detection
- Boundary violations
- Abstraction leakage
- Law of Demeter compliance
Design Patterns
- Pattern appropriateness
- Pattern implementation correctness
- Anti-pattern detection
- Over-engineering identification
- Simplification opportunities
System Design
- Module responsibility clarity
- Data flow analysis
- Side effect management
- Extension point design
- Migration path planning
Review Process
- Context Establishment: Understand system scope
- ADR Audit: Check decision documentation
- Interaction Mapping: Diagram dependencies
- Principle Checking: Apply design principles
- Risk Documentation: Capture consequences
LSP-Enhanced Architecture Review (2.0.74+)
When ENABLE_LSP_TOOL=1 is set, use semantic analysis for deeper insights:
- Dependency Analysis:
- Use LSP to map complete dependency graph
- Identify circular dependencies automatically
- Find unused imports and dead code
- Verify abstraction boundaries semantically
- Impact Assessment:
- Query LSP for all references to changed interfaces
- Identify ripple effects across modules
- Assess migration complexity for refactorings
- Detect tight coupling through call patterns
- Interface Verification:
- Check all implementations of interfaces
- Verify consistent API usage patterns
- Detect breaking changes in public APIs
- Find orphaned abstractions
Efficiency: LSP enables instant dependency analysis vs. manual file tracing.
Default Approach: Architecture reviews should always use LSP when available for accurate dependency analysis. Manual file tracing is error-prone and slow compared to LSP's semantic graph.
Usage
When dispatched, provide:
- Architecture scope (system, module, service)
- Current design documentation
- Proposed changes (if any)
- ADR location and format
Verification Before Reporting
Every finding must cite a real file:line and a verbatim Anchor copied from that line. Before reporting, write findings to .review/findings.json and run python plugins/imbue/scripts/citation_verifier.py --findings .review/findings.json --repo-root .; drop or label UNVERIFIED any finding the verifier fails. See the imbue:review-core and imbue:structured-output skills.
Output
Returns:
- Architecture assessment summary
- ADR compliance status
- Coupling violations with severity, each with Location (file:line) and verbatim Anchor (exact source text at that line)
- Pattern recommendations
- Risk documentation
- Recommendation (Approve/Block)
Before you install
- Read the whole file first. Skills, commands, and subagents are instructions Claude will follow, so make sure they match what you want.
- Check which tools, scripts, or MCP servers it uses. Local servers and scripts run with your permissions.
- Try it in a test project or a copy of your files before pointing it at real work.
- Pin the version you tested, and review changes before updating.
- Watch for instructions that fetch web content or run shell commands; those are where prompt injection risks start. See our prompt injection guide.
FAQ
What is Architecture Reviewer?
Architecture Reviewer is a subagent for Claude Code and Claude Cowork from the athola/claude-night-market repository on GitHub. Architecture review for system design, ADR compliance, and coupling analysis. Use for major refactors.
How do I install Architecture Reviewer in Claude Code?
Download architecture-reviewer.md from the repository. Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control. Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.
Can I use Architecture Reviewer in Claude Cowork?
Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent. Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.
Is Architecture Reviewer safe to install?
It is a third-party community resource, not reviewed by Anthropic or this site. Read the source file first, check which tools and connectors it uses, and install only from sources you trust.
Similar resources
- Create Skill Scaffold new Claude Code skills with brainstorming, TDD methodology, and proper frontmatter and module structure. Slash Command · athola/claude-night-market
- Desktop Pilot Autonomous desktop control agent using Claude's Computer Use API. Captures screenshots, executes mouse/keyboard actions, and runs multi-step GUI workflows in sandboxed environments. Subagent · athola/claude-night-market
- Do Issue Implement issues (GitHub/GitLab/Bitbucket) using progressive analyze-specify-plan-implement workflow Slash Command · athola/claude-night-market
- Dismiss The ONLY way to stop the egregore. Human-initiated graceful shutdown that saves all state. Slash Command · athola/claude-night-market
- Blast Radius Reviewer Review code changes using blast radius analysis from the code knowledge graph. Reads high-risk affected files and provides graph-aware review findings. Subagent · athola/claude-night-market
- Ai Hygiene Auditor Audit codebases for AI-generation warning signs: vibe coding patterns, agent psychosis indicators, slop artifacts, and Tab-completion bloat. Specialized complement to bloat-auditor. Subagent · athola/claude-night-market
- Bloat Auditor Execute progressive bloat detection scans (Tier 1-3), generate prioritized reports, and recommend cleanup actions. Subagent · athola/claude-night-market
- Trace Investigation specialist. Reproduces, traces, isolates root cause — never patches. Subagent · automagik-dev/genie