Claude Code Plugins Marketplace

A Claude Code plugin marketplace. Plugins cover the full development lifecycle: release automation, Home Assistant integration dev, Qt UI development, deep web research, and three-layer documentation propagation.

Table of Contents

• Claude Code Plugins Marketplace
  • Table of Contents
    • Staying Up to Date
  • Available Plugins
  • Principles
    • Home Assistant Dev
    • Release Pipeline
    • Qt Suite
    • qdev
    • Test Driver
    • Up Docs
  • Testing & Validation
  • Plugin Development
    • Quick Start
    • Documentation
  • Repository Structure
  • Contributing
  • License

Add this marketplace to your Claude Code installation:

/plugin marketplace add L3DigitalNet/Claude-Code-Plugins

Or using the full URL:

/plugin marketplace add https://github.com/L3DigitalNet/Claude-Code-Plugins.git

Then install individual plugins:

/plugin install release-pipeline@l3digitalnet-plugins

Staying Up to Date

Auto-update keeps plugins current automatically. To verify it's enabled:

1. Run /plugin in Claude Code
2. Go to the Marketplaces tab
3. Select l3digitalnet-plugins
4. Look for Disable auto-update (meaning it's already on)

When auto-update is enabled, Claude Code refreshes the marketplace catalog and updates installed plugins at the start of each session.

Manual update if you prefer to control when updates happen:

# Refresh the marketplace catalog
/plugin marketplace update l3digitalnet-plugins

# Then update individual plugins via /plugin → Installed tab → Update now

Available Plugins

Plugin	Type	Command	Description
Home Assistant Dev	Commands + Skills + MCP	varies	Full HA integration development toolkit with 27 skills
qdev	Commands + Agents	/research	Deep web research for development decisions: dual-source sweeps with Context7 docs gating, persisted as cited reports under docs/research/
Qt Suite	MCP + Commands + Skills + Agents	/qt-suite:scaffold, /qt-suite:coverage, /qt-suite:visual	Complete Qt development and testing toolkit: proactive agents, 16 skills, scaffolding, and headless GUI testing
Release Pipeline	Commands + Skills	/release	Semver releases with pre-flight checks and changelog generation
Test Driver	Commands + Skills	/test-driver:analyze, /test-driver:status	Proactive testing via gap analysis, convergence loops, and persistent status tracking
Up Docs	Skills + Agents	/up-docs:repo, /up-docs:wiki, /up-docs:notion, /up-docs:all, /up-docs:drift	Update documentation across three layers via dispatched sub-agents (Haiku repo/Notion propagators + Sonnet wiki propagator & drift auditor) from session context, plus full infrastructure drift analysis

Principles

These principles apply across all plugins in this collection. Individual plugins may define additional principles scoped to their domain.

[P1] Act on Intent: Invoking a command is consent to its implied scope. When intent is ambiguous, clarify scope before executing, not after. When intent is clear, execute without narration or confirmation of the obvious. A confirmation gate is warranted only when an operation is both truly irreversible and its scope materially exceeds what the invocation implies, not for operations that are merely large or look dangerous. If scope materially changes mid-execution, stop and re-confirm. Routine friction is a tax, not a safeguard.

[P2] Scope Fidelity: Execute the full scope of what was asked, completely, without routine sub-task confirmation gates. When a sub-task's scope is genuinely ambiguous, clarify before proceeding rather than assuming. Do not act beyond the declared scope; surface only material out-of-scope findings as notes; routine adjacent observations are not worth raising. Scope undershoot triggers additional iteration; scope overshoot violates the consent established at invocation.

[P3] Succeed Quietly, Fail Transparently: Lead with findings, not intent or preamble. Output results, not narration. Emit a compact factual summary at task completion, not after every step. On transient or infrastructure errors, retry silently. On critical failures, stop immediately and surface the complete failure (raw output included) with a recovery plan; do not attempt autonomous workarounds.

[P4] Use the Full Toolkit: When interaction is required, prefer structured choices over open-ended prompts; bound the user's decision space before presenting it. Use Claude Code's interaction primitives: bounded choices (AskUserQuestion), markdown previews for side-by-side comparisons, multi-select for non-exclusive options.

[P5] Convergence is the Contract: Iterative work defines completion as a measurable criterion (set by the plugin, the user, or collaboratively) and drives toward it without check-ins. Proceed quietly when converging normally. Surface immediately if progress stalls or regresses unexpectedly. If the cycle begins oscillating (making and undoing the same changes repeatedly), flag the pattern and stop rather than continuing. Stop only when the criterion is met, oscillation is detected, or the user intervenes.

[P6] Composable, Focused Units: Every plugin component (command, skill, hook) does one thing and is independently useful. Complex workflows emerge from combining atomic units at runtime; orchestration is assembled from the outside, not baked in.

---

Home Assistant Dev

Home Assistant integration development toolkit: 27 AI skills, an MCP server for live HA connections, automated validation, example integrations, and project templates.

Features:

• 27 context-aware skills covering architecture, config flows, coordinators, entities, testing, and more
• 3 specialized agents (development, review, debugging)
• MCP server with 12 tools for live HA connection and documentation search
• 5 validation scripts with PostToolUse hook enforcement
• 3 example integrations (Bronze/Silver/Gold tier)
• 9 project templates for CI/CD, testing, and documentation
• Full Integration Quality Scale coverage (all 52 rules)

Install:

/plugin install home-assistant-dev@l3digitalnet-plugins

Learn more: plugins/home-assistant-dev/README.md

---

Release Pipeline

Autonomous release pipeline: quick merge or full semver release with parallel pre-flight checks, changelog generation, and GitHub release creation.

Features:

• Two modes: quick merge (testing → main) or full versioned release
• Parallel pre-flight agents (test runner, docs auditor, git validator)
• Automatic changelog generation from conventional commits
• Version bumping across Python, Node.js, Rust, and plugin manifests
• GitHub release creation with release notes
• Human-in-the-loop approval gates at critical stages
• Fail-fast with rollback guidance on errors

Install:

/plugin install release-pipeline@l3digitalnet-plugins

Learn more: plugins/release-pipeline/README.md

---

Qt Suite

Complete Qt development and testing toolkit: proactive specialist agents, 16 domain skills, scaffolding commands, and headless GUI testing via the bundled Qt Pilot MCP server. Covers PySide6, PyQt6, and C++/Qt.

Features:

• 4 proactive agents: development specialist, debugger, code reviewer, and UX advisor
• 16 context-aware skills covering signals/slots, layouts, Model/View, threading, QML, styling, and more
• /qt-suite:scaffold: generates a complete PySide6 project with pyproject.toml, src layout, and test config
• /qt-suite:generate: scans codebase and generates unit tests for untested files
• /qt-suite:coverage: gcov/lcov (C++) or coverage.py (Python) report with gap-targeted test generation
• /qt-suite:visual: launches app headlessly via Xvfb, drives UI via the bundled Qt Pilot MCP server

Install:

/plugin install qt-suite@l3digitalnet-plugins

Learn more: plugins/qt-suite/README.md

---

qdev

Deep web research: a single user-initiated command that sweeps current sources before you design or build.

Features:

• /research: dual-source sweep (Tavily-first recall, Brave/Serper cross-checks, Context7 docs gating, footgun corroboration) persisted as a cited, frontmatter-indexed report under docs/research/

Install:

/plugin install qdev@l3digitalnet-plugins

Learn more: plugins/qdev/README.md

---

Test Driver

Proactive testing awareness and gap filling: always-on testing mindset that suggests gap analysis at natural breakpoints, finds missing tests across six categories, and iterates through a convergence loop until everything passes.

Features:

• Always-on testing-mindset skill triggers at natural breakpoints
• Six-category gap analysis (unit, integration, e2e, contract, security, UI)
• Convergence loop with oscillation detection
• Stack profiles for FastAPI, Django, PySide6, Home Assistant, and Swift
• Persistent test status tracking across sessions

Install:

/plugin install test-driver@l3digitalnet-plugins

Learn more: plugins/test-driver/README.md

---

Up Docs

Three-layer documentation updater via sub-agent dispatch: infers what changed during a session and updates repo docs, llm-wiki, and Notion at the right level of detail for each layer. Also provides comprehensive drift analysis that SSHes into live infrastructure.

Features:

• Parallel sub-agent architecture: two Haiku propagators (repo, notion) + one Sonnet (wiki) run in isolated context windows for cost efficiency, while Sonnet audit ensures drift detection quality
• /up-docs:repo, /up-docs:wiki, /up-docs:notion, /up-docs:all: dispatch targeted propagators from session context
• /up-docs:drift: four-phase convergence loop that gathers live server state via SSH, syncs llm-wiki, resolves internal contradictions, verifies and enriches links, then updates Notion
• Wall-clock time to completion reduced to max(repo, wiki, notion) via parallel dispatch; sequential drift audit phases protect data integrity

Install:

/plugin install up-docs@l3digitalnet-plugins

Learn more: plugins/up-docs/README.md

---

Testing & Validation

The marketplace standardizes test frameworks per language — bats for bash, pytest for Python, Jest for TypeScript. See docs/handoff/conventions.md (TEST-001) for the canonical frameworks and per-language naming conventions.

Quick reference:

# Bash plugins
cd plugins/release-pipeline && ./tests/run-bats.sh

# Python plugins
pytest plugins/home-assistant-dev/tests/ -m unit

# Marketplace schema validation (always run before merging to main)
./scripts/validate-marketplace.sh

Each plugin's tests live under its own plugins/<plugin>/tests/ directory.

Plugin Development

This repository also serves as a development workspace for creating new plugins. See CLAUDE.md for architectural guidance and docs/ for the full reference.

Quick Start

1. Create a new plugin:

   mkdir -p plugins/my-plugin/.claude-plugin
   cd plugins/my-plugin

2. Add manifest:

   { "name": "my-plugin", "version": "0.1.0", "description": "Plugin description" }

3. Test locally:

   claude --plugin-dir ./plugins/my-plugin

4. Add to marketplace catalog (.claude-plugin/marketplace.json)

Documentation

• docs/plugins.md - Plugin development guide
• docs/plugin-marketplaces.md - Marketplace creation
• docs/plugins-reference.md - Technical reference
• docs/skills.md - Creating AI-invoked skills
• docs/sub-agents.md - Custom agent definitions
• docs/hooks.md - Lifecycle event handlers
• docs/mcp.md - MCP server integration

Repository Structure

Claude-Code-Plugins/
├── .claude-plugin/
│   └── marketplace.json        # Marketplace catalog
├── plugins/                     # All plugin implementations (6 plugins)
│   ├── home-assistant-dev/      # Home Assistant integration dev toolkit
│   ├── qdev/                    # Deep web research (research sweeps via qdev-researcher)
│   ├── qt-suite/                # Qt development and testing toolkit (agents, skills, MCP)
│   ├── release-pipeline/        # Autonomous release pipeline
│   ├── test-driver/             # Proactive testing via gap analysis and convergence
│   └── up-docs/                 # Three-layer documentation updater (repo, wiki, Notion)
├── scripts/
│   └── validate-marketplace.sh  # Marketplace validation
├── docs/                        # Comprehensive documentation
├── CLAUDE.md                    # Development guidance for AI agents
├── BRANCH_PROTECTION.md         # Branch protection and workflow guide
└── README.md                    # This file

Contributing

To add a plugin to this marketplace:

1. Create plugin in plugins/ directory
2. Add entry to .claude-plugin/marketplace.json (version must match the plugin's own plugin.json)
3. Validate with ./scripts/validate-marketplace.sh
4. Commit directly to main and push
5. To publish a tagged release with GitHub release notes, run /release-pipeline:release and pick the appropriate mode

Branch workflow: Direct commit to main. There is no testing branch. Local pre-commit hooks (noreply email enforcement, marketplace validation) provide guardrails. See BRANCH_PROTECTION.md for full rules.

git pull origin main
# (make edits)
./scripts/validate-marketplace.sh
git add <specific files>
git commit -m "..."
git push origin main

License

MIT - See LICENSE file for details