Conway Method
Use when working in a project that has a `.agents/` directory or when the user asks about Conway Architecture, the conway-architecture plugin, domain agents, owned_paths enforcement, or how to coordinate a persistent subagent team that shadows the application's architecture
Install
/plugin marketplace add tilmon-engineering/claude-skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add tilmon-engineering/claude-skills --skill conway-methodgit clone --depth 1 https://github.com/tilmon-engineering/claude-skills.git /tmp/claude-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills/plugins/conway-architecture/skills/conway-method ~/.claude/skills/conway-methodPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Conway Architecture: the method ## Core idea Conway's Law observes that **a system's architecture tends to resemble the communication structure of the team that built it**. Usually treated as a warning. Here, it's used as a tool: design the team you want, and let the architecture follow. The team is a **persistent, domain-scoped subagent team**. Each domain owns a disjoint slice of the codebase, declared as path globs. The coordinator (the main session agent) routes work; the domain agents do the work. The boundaries are enforced by a `PreToolUse` hook, not just by convention. ## When to use - The project has (or should have) a `.agents/` directory with per-domain `AGENTS.md` files. - You are about to delegate work across domains and need to know who owns what. - The user mentions "Conway", "the team", "domain agents", or asks why a write was blocked. - You see hook denials mentioning "Conway:" — that's this method enforcing itself. Skip when: the project has no `.agents/` directory. The plugin is a no-op there. ## Vocabulary | Term | Meaning | |---|---| | **Coordinator** | The main session agent. Owns the roadmap, never edits domain-owned files. Identifiable in hook input by the *absence* of `agent_type`. | | **Domain agent** | A teammate spawned via `Agent(team_name=…, name=<domain>, …)`. The `name` matches a directory under `.agents/`. Identifiable in hook input by `agent_type == <domain>`. |
Details
- Skill name
conway-method- Repository
- tilmon-engineering/claude-skills
- Path
plugins/conway-architecture/skills/conway-method/SKILL.md- License
- MIT
- Repo updated
- 22 May 2026
FAQ
What does the Conway Method skill do?
Use when working in a project that has a `.agents/` directory or when the user asks about Conway Architecture, the conway-architecture plugin, domain agents, owned_paths enforcement, or how to coordinate a persistent subagent team that shadows the application's architecture
How do I install the Conway Method skill?
Claude Code plugin: /plugin marketplace add tilmon-engineering/claude-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Conway Method popular?
It lives in tilmon-engineering/claude-skills, which has 5 GitHub stars and 2 forks.
Related skills
Analyze Branch
5tilmon-engineering/claude-skills
Use when analyzing another branch's iteration journals to extract findings, decisions, and insights from divergent work
Branch Status
5tilmon-engineering/claude-skills
Use when user wants detailed status report for single autonomy branch including iteration timeline and metrics progression
Checkpoint Iteration
5tilmon-engineering/claude-skills
Use when saving current iteration progress mid-conversation, before context compaction, or at interim pause points
Smart Screenshot
5astoreyai/claude-skills
Intelligent screenshot and screen capture with OCR, markdown conversion, and annotation. Triggered by PrtSc key, captures screen regions, extracts text with OCR, converts to markdown using MarkItDown, saves with auto-formatting. Similar to Windows Snipping Tool with AI enhancements for text extraction and document processing.
Webmcpify
9sandbaseai/workbuddy-skill
Make a web app agent-ready — propose a WebMCP tool manifest, integrate, verify in a real browser, heal; unrelated code stays untouched. Use for "webmcpify", "add WebMCP", or "expose app actions to AI agents".
Text Metrics
5astoreyai/claude-skills
Probability-based text analysis providing GLTR token rank histograms, DetectGPT curvature probes, and Coh-Metrix-inspired cohesion metrics. Designed to compose with ai-check for comprehensive AI writing pattern detection.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.