Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By bdfinst
Orchestrate a multi-agent AI dev team in Claude Code to plan TDD implementations, generate code with quality gates, run specialized reviews (security, accessibility, performance, architecture), automate PRs, and handle DevOps/CI-CD setups via slash commands and personas like architect, SRE, and QA engineer.
npx claudepluginhub bdfinst/agentic-dev-team --plugin agentic-dev-teamInstall a Claude Code plugin and register it in settings.json so the full team can replicate the install. Use this whenever adding a new plugin to the project — it keeps settings.json in sync with what is actually installed.
Scaffold a new review agent from a description or URL. Use this whenever the user wants to add a new review agent, detect a new category of code issue, or says things like "add an agent for X", "create a reviewer for Y", "I want to check for Z in code reviews". Also use when given a URL to a coding standard or best-practices guide that should become a review agent.
Audit code-review agents, skills, and hooks for structural compliance. Use this when adding or modifying any agent, skill, or hook file, or for a periodic health check of the toolkit. Trigger phrases: "audit the agents", "check compliance", "validate the skills", "are the agents correct", or any time agent/skill files change.
Run eval fixtures against review agents and grade results. Use this after adding or modifying a review agent, to validate detection accuracy, or when the user says "run the evals", "test the agents", "check for regressions", or "how accurate is the agent".
Remove an agent from the system — deletes the agent file, cleans up all registry entries, removes cross-references, and updates documentation. Use when the user says "remove the X agent", "delete X-review", "retire the X role", or "we no longer need X". Handles both team agents and review agents. Always confirms before deleting.
WCAG 2.1 AA compliance, semantic HTML, ARIA, keyboard navigation, focus management
Creates and manages Architecture Decision Records (ADRs) with a decision framework for when to create one
Architectural alignment — ADR compliance, layer boundary violations, dependency direction, pattern consistency
System design, architecture definition, and technical decision oversight
CLAUDE.md completeness, rules, skills, and path accuracy
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
End-to-end development workflow: design → draft-plan → orchestrate → review → pr-create → pr-review → pr-merge
Self-orchestrating multi-agent development system — 8 specialized AI agents, parallel quality gates, and automated workflows. You say WHAT, the AI decides HOW.
A curated set of skills for each stage of development — propose, spec, design, plan, implement, ship.
AI-powered development workflow automation - Phase-based planning, implementation orchestration, preflight code quality checks with security scanning, ship-it workflow, and development principles generator for CLAUDE.md
Virtual development team: TDD, debugging, code review, backlog management, and proven workflow patterns
Implementation planning, execution, and PR creation workflows with multi-agent collaboration
A Claude Code plugin that adds a full persona-driven AI development team to any project. The Orchestrator routes tasks to specialized agents, inline review checkpoints catch quality issues during implementation, and skills provide reusable knowledge modules that any agent can draw on.
Four commands drive feature development from idea to pull request:
/specs → /plan → /build → /pr
| Step | Command | What it does |
|---|---|---|
| 1. Specify | /specs | Collaborate on four artifacts: Intent, BDD/Gherkin scenarios, Architecture notes, Acceptance Criteria. A consistency gate must pass before moving on. Skip for bug fixes, refactors, or trivial changes. |
| 2. Plan | /plan | Create a step-by-step TDD implementation plan. Checks for spec artifacts first — if none exist, asks whether to continue or run /specs. Human approves before any code is written. |
| 3. Build | /build | Execute the approved plan. Each step follows RED-GREEN-REFACTOR with inline review checkpoints (spec-compliance first, then quality agents). Produces verification evidence. |
| 4. Ship | /pr | Run quality gates (tests, typecheck, lint, code review) and create a pull request. |
Each step produces artifacts the next step consumes. Human review gates sit between each transition.
flowchart LR
S["/specs\n4 artifacts"] -->|consistency gate| P["/plan\nTDD steps"]
P -->|human approval| B["/build\nRED-GREEN-REFACTOR"]
B -->|code review| PR["/pr\nquality gates"]
For bug fixes or simple tasks, skip /specs and start at /plan or go straight to implementation. The orchestrator routes trivially when the full workflow isn't needed.
| Command | When to use |
|---|---|
/code-review | Run all review agents against changed files (also runs as part of /build) |
/continue | Resume an in-progress build or plan across sessions |
/browse | Visual QA via Playwright |
/careful / /freeze / /guard | Safety modes for production-critical sessions |
Every git commit is automatically gated by /code-review --changed. A PreToolUse hook detects commit attempts and blocks them until a passing review exists for the exact set of staged files.
Flow: attempt commit → hook blocks → Claude runs /code-review --changed → if pass/warn, a .review-passed gate file is written → next commit attempt succeeds.
Bypass: git commit --no-verify skips the review gate.
Team agents define roles (persona, behavior, collaboration). Review agents check work quality in real time. Skills define knowledge (patterns, guidelines, procedures). Slash commands invoke agents and skills directly. The Orchestrator controls task routing, model selection, and the inline review feedback loop.
For complex tasks where the orchestrator manages the full lifecycle, every non-trivial task follows Research → Plan → Implement with human review gates between phases:
docs/specs/) with problem statement, alternatives, and scope boundariesflowchart TD
U([User Request]) --> O[Orchestrator]
subgraph "Phase 1 — Research"
O --> SP["/specs\nIntent · BDD · Architecture · Criteria"]
SP --> DD["Design Doc\ndocs/specs/"]
end
DD --> HG1([Human Gate — approve spec + design])
subgraph "Phase 2 — Plan"
HG1 --> PL["/plan\nTDD steps · complexity tiers · acceptance criteria"]
PL --> PR["Plan Reviewer\nprompts/plan-reviewer.md"]
end
PR --> HG2([Human Gate — approve plan])
subgraph "Phase 3 — Implement (/build)"
HG2 --> CV["Verify criteria testability\n(sprint contract gate)"]
CV --> NEXT{"Next step?"}
NEXT -->|more steps| IM["TDD Loop\nRED → GREEN → REFACTOR"]