Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By qazuor
End-to-end project planning with specs, task decomposition, dependency graphs, quality gates, autonomous loops, and guardrails
npx claudepluginhub qazuor/claude-code-plugins --plugin task-masterCancel an active autonomous work loop and show summary
Start an autonomous work loop that processes tasks sequentially with quality gates
Create a standalone task directly without requiring a specification
Find and start the next available task based on dependencies and complexity
Re-plan tasks when requirements change - add, remove, modify, reorder, or split tasks safely
Generates functional specifications with user stories, BDD acceptance criteria, UX considerations, and scope definition
Decomposes specifications into implementable atomic tasks with dependencies, phases, and complexity scoring
Generates technical analysis including architecture design, data model changes, API design, risk assessment, and performance considerations
Analyzes tasks and assigns complexity scores (1-10) with a hard ceiling of 4 for atomic tasks, flagging anything above for further decomposition
Computes dependency graphs for task sets, validates for cycles, identifies critical paths, parallel tracks, and execution levels
Detects overlaps between a new requirement and existing specs/tasks to prevent duplicate work and identify related efforts
Runs lint, typecheck, and test quality checks before marking a task as completed, updating state and regenerating progress reports
Takes Plan Mode analysis output and generates formal spec documents with proper IDs, metadata, and template-based content
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.
Spec-Driven Development: specifications compiled into executable, verifiable behavior
Installer for TÂCHES' Get-Shit-Done workflow system - meta-prompting, context engineering and spec-driven development
AI-native task management system. Track progress, coordinate multi-agent work, manage hierarchical tasks with dependencies, sprints, acceptance criteria, and a web dashboard.
Turn broad Codex and Claude Code work into pressured /goal runs with oracles, local boards, receipts, and verification.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Synchronize agents, skills, commands, docs, and templates from claude-code-knowledge repository to your projects
Project initialization with CLAUDE.md generation, merge support, settings templates, and full setup orchestration
Pre-configured MCP server definitions for 30 services including databases, APIs, browser testing, and more
Auto-learn and synchronize Claude Code permissions across all projects
Session management tools including diary entries, reflections, and claude-mem watchdog
A curated collection of plugins for Claude Code (Anthropic's official CLI tool). These plugins enhance your Claude Code experience with notifications, task management, permission synchronization, and more.
In Claude Code, run /plugin and select Add marketplace:
qazuor/claude-code-plugins
After adding the marketplace, install the plugins you want:
/plugin → Install plugin → Select from qazuor-claude-code-plugins
Recommended minimum:
notifications - Audio feedback when Claude finishespermission-sync - Keep permissions consistent across projectstask-master - If you work with specs and tasksSome plugins register hooks that only activate on session start. Restart Claude Code after installing.
| Plugin | Purpose | Auto-runs |
|---|---|---|
| notifications | Desktop notifications + audio beeps | On events |
| task-master | Specs, tasks, progress tracking | SessionStart |
| mcp-servers | 30 pre-configured MCP servers | Manual |
| permission-sync | Sync permissions across projects | SessionStart |
| knowledge-sync | Sync agents/skills from external repo | SessionStart |
| session-tools | Diary entries before compaction | PreCompact |
| claude-initializer | Initialize new projects | Manual |
Purpose: Audio and visual feedback when Claude Code needs your attention.
What it does:
What it doesn't do:
Hooks:
| Event | Action |
|---|---|
Notification | Desktop notification + TTS |
Stop | 1000Hz beep (0.2s) |
SubagentStop | 800Hz beep (0.1s) |
Requirements: paplay (Linux), afplay (macOS), or PowerShell (Windows/WSL)
Purpose: Specification-driven development with task decomposition, progress tracking, autonomous loops, and guardrails.
What it does:
What it doesn't do:
Commands:
| Command | Description |
|---|---|
/spec | Create a specification from requirements |
/tasks | Show task dashboard with progress |
/next-task | Get the next available task |
/new-task | Create a standalone task (no spec needed) |
/task-status | Detailed progress report |
/replan | Modify tasks when requirements change |
/auto-loop | Start autonomous task processing loop |
/auto-loop-cancel | Cancel an active autonomous loop |
Autonomous Loop:
The /auto-loop command starts a controlled loop that processes tasks one after another:
Guardrails:
Learned constraints stored in .claude/guardrails.md: