By GROOVIZ
An opinionated eight-skill methodology for running architecture investigations with two user-review gates, per-step verification, and a closing head-of-architecture review. Inspired by Matt Pocock's skills repo.
Relentlessly interview the user about an architecture investigation until scope, drivers, methodology, and deliverables are 100% concrete. MUST use AskUserQuestion for every question, multiple-choice only, with a recommended answer. Minimum six questions. Ends with explicit user confirmation. Use when the user starts an architecture investigation, audit, migration scoping, or feature deep-dive.
Final senior-architect review of a completed investigation. Reads the brief, the plan, the synthesis, a sample of per-item outputs, and the trace log; returns a written review that an engineering leader would sign — or send back. Use as the closing skill of an investigation, after `synthesize-investigation`.
Audit a freshly-written Investigation Brief for gaps, ambiguity, and weak definitions, then close every gap — by re-reading the grilling, re-reading the codebase, or grilling the user again via AskUserQuestion. Iterates until the brief is solid. Invoked by the `/architecture-investigation:investigate` orchestrator after `write-investigation-brief` and before the user reviews. Sets `status: reviewed` when done.
Audit a freshly-written investigation plan for gaps — vague verifications, missing schema rows, underweight phases, single-option strategies — then close every gap by re-reading the brief, re-reading the codebase, or grilling the user again via AskUserQuestion. Iterates until the plan is solid. Invoked by the `/architecture-investigation:investigate` orchestrator after `write-investigation-plan` and before the user reviews. Sets `status: reviewed` when done.
Peer-review a verified plan-step output from an engineering-judgment perspective. Reads the step's output, considers what an experienced architect would push back on, and returns improvement suggestions. Use after `verify-step` passes, before moving to the next step.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
This repository is a Claude Code marketplace that distributes a single plugin: architecture-investigation.
# Add the marketplace
/plugin marketplace add <owner>/<this-repo>
# Install the plugin
/plugin install architecture-investigation@architecture-harness
For local development:
/plugin marketplace add ./
/plugin install architecture-investigation@architecture-harness
Once installed, run:
/architecture-investigation:investigate [investigation-id]
| Plugin | What it does |
|---|---|
architecture-investigation | Eight-skill methodology for architecture investigations, with two user-review gates, dual-track per-step verification + peer review, and a plan-declared per-item output schema. |
See architecture-investigation/README.md for plugin-level usage details, or docs/design.md for the design rationale behind the eight-skill methodology.
architecture-harness/ # this repo = the marketplace
├── .claude-plugin/
│ └── marketplace.json # marketplace manifest
├── architecture-investigation/ # the plugin
│ ├── .claude-plugin/plugin.json
│ ├── commands/
│ ├── skills/
│ ├── CLAUDE.md
│ └── README.md
├── docs/
│ └── design.md # design document
├── README.md # this file
└── .gitignore
# from inside Claude Code, against this repo:
/plugin validate .
# against the plugin specifically:
/plugin validate ./architecture-investigation
Style and structure borrowed from Matt Pocock's skills repo.
npx claudepluginhub grooviz/aqzsedrf2 --plugin architecture-investigationAuthoring kit for LLM-judged, schema-validated, replayable code analyzers. Scaffolds a domain-specific Claude Code skill via an interactive `/new-analyzer` interview that drafts rules.md through a dispatched rule-author subagent, auto-generates one fixture stub per rule, optionally seeds a dev-team expected-entities oracle, and gates success on a post-stamp validator. Stamped analyzers record per-entity property bags (Phase D side-effect + Phase D.5 consolidation) and consult the oracle to backstop discovery with within-run framework expansion. Ships five specialist subagents, a companion `/expected-entities` command, and a generic runtime (schema validator, scaffold validator, snippet normalizer, fixture comparator, oracle comparator, override-replay engine).
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Write feature specs, plan roadmaps, and synthesize user research faster. Keep stakeholders updated and stay ahead of the competitive landscape.