Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By Oiranca
A robust, scalable Claude Code plugin providing a multi-agent development ecosystem with specialized agents, reusable skills, slash commands, and operational guardrails.
npx claudepluginhub oiranca/claude-dev-ecosystemFull analysis and improvement workflow for an existing repository. Runs stack detection, inventory, dependency audit, architecture review, targeted improvements, and QA.
Incremental migration workflow from React + Vite to Astro using islands architecture. Validates preconditions, inventories routes, plans architecture, migrates components in batches, and validates.
Workflow for a new or bootstrapped project with few source files. Runs stack detection, inventory, architecture planning, initial implementation, and QA.
Safe, behavior-preserving refactor workflow for a specific module or file set. Includes planning, scoped implementation, and validation.
Focused security audit workflow. Prioritizes sensitive surfaces first, avoids reproducing secrets, and returns a remediation-oriented report.
Full analysis and improvement workflow for an existing repository. Runs stack detection, inventory, dependency audit, architecture review, targeted improvements, and QA.
Incremental migration workflow from React + Vite to Astro using islands architecture. Validates preconditions, inventories routes, plans architecture, migrates components in batches, and validates.
Workflow for a new or bootstrapped project with few source files. Runs stack detection, inventory, architecture planning, initial implementation, and QA.
Safe, behavior-preserving refactor workflow for a specific module or file set. Includes planning, scoped implementation, and validation.
Focused security audit workflow. Prioritizes sensitive surfaces first, avoids reproducing secrets, and returns a remediation-oriented report.
Web accessibility patterns for WCAG 2.2 compliance including ARIA, keyboard navigation, screen readers, and testing
Adapt designs to work across different screen sizes, devices, contexts, or platforms. Implements breakpoints, fluid layouts, and touch targets. Use when the user mentions responsive design, mobile layouts, breakpoints, viewport adaptation, or cross-device compatibility.
Review a feature and enhance it with purposeful animations, micro-interactions, and motion effects that improve usability and delight. Use when the user mentions adding animation, transitions, micro-interactions, motion design, hover effects, or making the UI feel more alive.
Detect likely drift between documented architecture and the current repository structure, file boundaries, and dependency patterns.
Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity ratings and actionable plan. Use when the user wants an accessibility check, performance audit, or technical quality review.
Requires secrets
Needs API keys or credentials to function
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.
Complete project development toolkit: 23 agents, 23 slash commands, 29 lifecycle hooks, and 69 reusable skills for Claude Code workflows
Portable Development System — AI-assisted development methodology with skills for consistency and agents for scale.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams
Universal Claude Code workflow with specialized agents, skills, hooks, and output styles for any software project. Includes orchestrator, code-reviewer, debugger, docs-writer, security-auditor, refactorer, and test-architect agents.
Agent Teams orchestration, governance hooks, multi-AI review, memento skill intelligence, and project management skills for Claude Code
Harness for Claude Code — skills, /harness:* slash commands, persona subagents, lifecycle hooks, and MCP tools without per-repo `harness setup`. Sibling plugins exist for Cursor, Gemini CLI, and Codex.
A parallel multi-agent development environment for Claude Code CLI. Provides a formal Task State Engine, typed agent messaging, distributed locks, and a shared knowledge layer — enabling agents to work concurrently without a central bottleneck.
| Feature | v1 | v2 |
|---|---|---|
| Task coordination | Shared Task List (informal) | Formal Task State Engine (tasks.json) |
| Agent communication | Via product-manager | Direct typed messages (messages.jsonl) |
| Parallel execution | Sequential phases | Concurrent claiming with dependency graph |
| Lock management | Basic lock files | TTL-based locks with auto-eviction |
| Shared timeline | None | timeline.log with atomic appends |
| Reviewer actions | Report only | APPROVE / REQUEST_CHANGES / ESCALATE |
agents/ # 12 specialized agent role definitions
skills/ # 15 reusable skill definitions (SKILL.md per skill)
commands/ # 7 workflow orchestrations
settings.example.json # Example Claude Code settings (copy to ~/.claude/settings.json)
scripts/
pre-edit-check.sh # Edit safety gate (hook script)
validate-local.sh # Validation runner (lint/typecheck/test/build)
agent-runtime.py # Task State Engine CLI ← core of v2
reference/
ARCHITECTURE_V2.md # v2 architecture reference
TEAM_MANUAL.md # Multi-agent team manual
GUARDRAILS.md # Operational guardrails
GUARDRAILS_REFERENCE.md # Detailed guardrail explanations
BUDGETS.md # Skill budget tiers
USAGE.md # Usage guide
templates/
local-scaffold/ # Optional docs/ and mcp/ templates for new repos
.mcp.json # MCP server config (filesystem + github)
CLAUDE.md # Global ecosystem rules loaded by Claude Code
~/.claude/)Installed once. Applies to all repositories.
~/.claude/
├── agents/
├── skills/
├── commands/
├── settings.json # Hooks: PreToolUse, SessionStart, Stop
├── scripts/
│ ├── pre-edit-check.sh
│ ├── validate-local.sh
│ └── agent-runtime.py ← Task State Engine
├── templates/
├── reference/
└── CLAUDE.md
Created per repository. Keep docs/ in git; keep .agent-cache/ gitignored.
<repo>/
├── CLAUDE.md # Optional repo-specific overrides
├── .claude/
│ └── settings.local.json
├── docs/ # Knowledge layer (committable, optional)
│ ├── STACK_PROFILE.md
│ ├── INVENTORY.md
│ ├── ARCHITECTURE.md
│ ├── DECISIONS.md
│ ├── TASKS.md
│ └── REVIEWS.md
└── .agent-cache/ # Runtime state (gitignored)
├── tasks.json
├── messages.jsonl
├── timeline.log
├── locks/
└── repo-map.json
| Agent | Model | Role |
|---|---|---|
| product-manager | sonnet | Plans, scopes, creates task graph, monitors blockers |
| pr-comment-responder | sonnet | Specialized Team Lead for PR review cycles |
| Agent | Model | Role |
|---|---|---|
| context-manager | haiku | Repo discovery; scoped reading plans |
| stack-analyzer | haiku | Stack detection |
| repo-analyzer | haiku | Repository structural inventory |
| solution-architect | sonnet | Architecture planning |
| Agent | Model | Role |
|---|---|---|
| software-engineer | sonnet | Feature implementation |
| migration-engineer | sonnet | Framework migrations |
| devops-engineer | sonnet | Infrastructure review |
| Agent | Model | Role |
|---|---|---|
| qa-engineer | haiku | Quality validation |
| security-reviewer | sonnet | Security audit |
| tech-writer | haiku | Documentation maintenance |
The Task State Engine is the coordination backbone. All agents use it.
# Create tasks (Team Lead)
python ~/.claude/scripts/agent-runtime.py task create \
--title "Detect stack" --owner stack-analyzer --priority high \
--inputs "Write to docs/STACK_PROFILE.md"
# Claim and work (Execution agents)
python ~/.claude/scripts/agent-runtime.py task list --status pending
python ~/.claude/scripts/agent-runtime.py task claim --id <id> --owner software-engineer
python ~/.claude/scripts/agent-runtime.py task update --id <id> --status running
python ~/.claude/scripts/agent-runtime.py task complete --id <id> --outputs "Modified: src/foo.ts"
# Communicate (all agents)
python ~/.claude/scripts/agent-runtime.py message send \
--from software-engineer --to qa-engineer \
--task-id <id> --type handoff \
--summary "Implementation complete. Validate lint, types, tests." \
--files "src/foo.ts,src/bar.ts"
python ~/.claude/scripts/agent-runtime.py message inbox --agent qa-engineer --unread