Claude Code plugin
Specmanager
Spec-driven development plugin for Claude Code. SpecManager gives you an agile AI team that takes a feature from interview → PRD → Architecture → Design → Plan → Build → Walkthrough.
Install
/plugin marketplace add joanseg/specmanagerFound in the project README. Check it for setup steps after installing.
git clone https://github.com/joanseg/specmanager.gitFrom the README
SpecManager Your AI product team, as a Claude Code plugin. As a non technical founder or product vibe coder you already have the builder — Claude Code. What you *don't* have is the rest of the team: the product manager who pins down the problem, the architect who fits it to your codebase, the planner who breaks it into shippable pieces, the reviewer who checks the work against the plan. SpecManager is that team. It gives you a set of specialised AI agents organised like an agile squad: the agents draft, you review, edit, and approve, and you decide what moves forward. It's spec-driven development you can actually see — a kanban board running on your own machine, one row per feature, one column per stage: PRD → Architecture → (Design) → Plan → Build → Walkthroughs, plus a research rail before any feature exists. Everything the team produces is plain markdown saved inside your project, so it reads like documents, not code, and your version history keeps every draft. It runs entirely on your own machine — single-user, private, no login. specmanager.org · MIT licensed Your team Each role is its own specialised agent — not one mega-prompt wearing different hats: - Researcher *(optional)* — works before there's a feature to build: a short brief, market research, Docs (wireframes, one-pagers, presentations), a user-research plan and a marketing plan, then findings from your interviews and a strategy whose build list becomes features. - Product manager — drafts the PRD, and can run an optional pre-PRD interview that pulls the idea out of your head and stress-tests it before you write anything down. - Architect — reads your actual codebase (its structure, conventions, the pieces already there) before designing, so "we'll add X" always points at something real instead of…
How plugins work
- A plugin bundles slash commands, subagents, skills, hooks, MCP servers and status lines so they install together.
- Add the marketplace inside Claude Code with /plugin marketplace add joanseg/specmanager, then run /plugin to pick what to install.
- You can turn a plugin off again from /plugin without deleting anything.
Details
- Repository
- joanseg/specmanager
- Website
- www.specmanager.org
- Language
- TypeScript
- License
- MIT
- Last updated
- 3 November 2026
- Topics
- ai-agentsai-codingclaude-codeclaude-code-pluginmcpspec-driven-developmentvibe-coding
FAQ
What is Specmanager?
Spec-driven development plugin for Claude Code. SpecManager gives you an agile AI team that takes a feature from interview → PRD → Architecture → Design → Plan → Build → Walkthrough. It is a plugin for Claude Code, from joanseg/specmanager on GitHub.
How do I install Specmanager?
Run /plugin marketplace add joanseg/specmanager inside Claude Code, then follow the setup in the README.
Is Specmanager popular?
joanseg/specmanager has 26 GitHub stars and 0 forks. It was last updated on 3 November 2026.
More plugins & marketplaces
Axme Code
14AxmeAI/axme-code
Persistent memory, architectural decision enforcement, pre-execution safety hooks, and session handoff for Claude Code. MCP server plugin for AI coding agents.
Review Squad
132389-research/review-squad
Dispatch panels of specialized subagents to review projects — expert audits, first-impression personas, task-completion flows, and pedantic nitpicks
Aiwkf
27k8adev/aiwkf
Agents, skills and plugins I use daily to code with AI. A living Claude Code marketplace, also compatible with Codex.
Spotnote
25harukikuri/spotnote
Vite plugin for UI feedback to coding agents.
Harness AI Kit
25seed-forge/harness-ai-kit
Package manager for AI agent assets — 42 skills, 5 CLIs, 1 plugin. Skills for AI/LLM agent engineering, eval-driven dev, spec-driven dev, database (MySQL/PG/Redis/Kafka/Mongo/Oracle/NL2SQL), K8s/Docker diagnostics, infra ops (Dify/Nexus/Harbor/SonarQube) & docs/patent. Runtimes: Codex, Claude Code, Cursor, Kiro, DSH.
Tailtest
12avansaber/tailtest
tailtest auto-generates and runs tests for every file your AI agent edits in Claude Code. R1-R15 rule layer + adversarial test mode (V13).
Browse the most popular Claude Code mods, all plugins & marketplaces, or the skills and MCP server directories.