Claude Code Marketplace

A Claude Code plugin marketplace providing development tools and productivity utilities.

Overview

This marketplace contains plugins for facilitated planning (CPM), note searching, PHP code intelligence, JavaScript/TypeScript code simplification, and Filament v5 admin mockups. All tools are designed to work seamlessly with Claude Code.

Installation

Install from Marketplace (inside Claude Code)

# Install the marketplace
/plugin marketplace add ninthspace/claude-code-marketplace

# Install individual plugins
/plugin install noteplan@claude-code-marketplace
/plugin install php-lsp@claude-code-marketplace
/plugin install cpm@claude-code-marketplace
/plugin install js-simplifier@claude-code-marketplace
/plugin install filament-mockup@claude-code-marketplace

Also needed for /cpm:ralph — the ralph-loop fork

CPM's autonomous loop (/cpm:ralph) does not implement the loop itself. It writes a state file and relies on a Stop hook supplied by a separate plugin. Install ours:

/plugin marketplace add ninthspace/ralph-loop
/plugin install ralph-loop@ninthspace-ralph

This is a fork of Anthropic's ralph-loop (Apache-2.0), which is itself the maintained line descended from Daisy Hollman's ralph-wiggum. The fork's first commit is the upstream plugin unmodified, so every change is visible as a diff against it.

One behavioural change, and it is the reason to prefer the fork: the Stop hook no longer deletes the loop's state file when it cannot read the transcript. Upstream treats a missing transcript, a failed grep for assistant records, and a jq parse error as reasons to end the run — and the state file is the loop, so ending it that way exits 0 with no promise and no state file, which is indistinguishable from a completed run. The fork continues on all three: failing to read a turn's output means no promise this iteration, not the work is finished. An unattended overnight run is precisely where that difference is invisible and expensive.

CPM works with any of the three — cpm/hooks/lib/ralph-hook-probe.sh detects the hook whichever plugin provides it, and warns when none is present. The upstream plugins remain usable; the fork is what /cpm:ralph is developed and tested against.

Do not enable more than one at a time. All three register a Stop hook, both fire on the same session, and the state file only has to be deleted by one of them for the loop to end.

Available Plugins

NotePlan Search (v1.0.0)

Search and query NotePlan notes from Claude Code

A skill for searching NotePlan content across:

• Notes folder - Standalone notes
• Calendar folder - Daily/weekly/monthly notes
• Spaces - Team/shared notes (SQLite database)
• iCloud - If syncing via iCloud Drive

Results are sorted by most recently modified first.

Quick Start:

# Search for a term
/noteplan coffee

# List all Spaces notes
/noteplan --list --spaces

# Fetch full note by ID
/noteplan --get UUID

# Search with date filters
/noteplan meeting --after 2025-01-01

# Natural language queries
/noteplan find me everything about project planning

Key Features:

• Full-text search across all NotePlan sources
• Date filtering (--after, --before)
• JSON output for AI tools
• Direct noteplan:// URLs to open notes in the app
• Excludes @Templates, @Trash, @Archive by default (use --all to include)

Requirements:

• macOS with NotePlan 3 installed
• Python 3

View full documentation

PHP LSP (v1.0.0)

PHP semantic code intelligence for Claude Code

Adds 24 LSP tools to Claude Code for PHP files via intelephense and the lsp-mcp-server bridge.

Capabilities:

• Go-to-definition, find references, find implementations
• Hover info (type signatures, documentation)
• Code completion and signature help
• Diagnostics (errors, warnings) per file and project-wide
• Safe rename across entire codebase
• Code actions (quick fixes, refactoring)
• Call hierarchy and type hierarchy
• File analysis (imports, exports, related files)
• Document formatting

Quick Start:

# One-time setup (installs intelephense + lsp-mcp-server, configures project)
/php-lsp:setup

# Restart Claude Code — LSP auto-starts on first use

# Check everything is working
/php-lsp:status

Requirements:

• Node.js >= 18
• Git

View full documentation

Claude Planning Method (v3.11.0)

Facilitated planning with multi-perspective party mode and focused consultation for Claude Code

Structured discovery, product ideation, architecture exploration, specification, work breakdown, task execution, retrospectives, and course correction through guided conversation. Includes party mode — a multi-agent discussion where named specialist personas (PM, Architect, Developer, UX Designer, QA, DevOps, Tech Writer, Scrum Master) debate trade-offs and surface blind spots — and consult mode for focused one-to-one expert dialogue with dynamic membership. Inspired by the BMAD-METHOD.

v3 is tuned for Opus 5 and later: all skills use positive-voice instructions, explicit stop criteria, outcome-oriented procedural guidance, and a reduced token footprint.

Twenty-two skills forming a pipeline:

Quick Start:

# Brainstorm with your team of agent personas
/cpm:party should we use a monorepo or separate repos?

# Focused consultation with one expert
/cpm:consult Margot

# Full pipeline: discover → brief → architect → spec → epics → do → retro
/cpm:discover build a customer portal for our booking system
/cpm:brief docs/plans/01-plan-customer-portal.md
/cpm:architect docs/briefs/01-brief-customer-portal.md
/cpm:spec docs/briefs/01-brief-customer-portal.md
/cpm:epics docs/specifications/01-spec-customer-portal.md
/cpm:do
/cpm:retro

# Import reference docs for skills to use as context
/cpm:library docs/architecture-decisions.md

# Review planning artifacts before or after execution
/cpm:review docs/epics/01-epic-customer-portal.md

# Course correct mid-flow
/cpm:pivot docs/specifications/01-spec-customer-portal.md

# Transform artifacts for stakeholders
/cpm:present docs/specifications/01-spec-customer-portal.md

# Explore and customise templates
/cpm:templates preview brief

# Clean up completed artefacts
/cpm:archive

# Hand a whole spec to the autonomous loop — epics first, then the work
/cpm:ralph docs/specifications/01-spec-customer-portal.md

# Small change? Skip the full pipeline
/cpm:quick add a --verbose flag to the deploy script

# Check project status and get next-step recommendations
/cpm:status

# Clean up leftover session-state files on demand
/cpm:clean

# Register a published artifact against the work that produced it
/cpm:artifact https://claude.ai/code/artifact/... auth flow explorer, from spec 12

# Or jump to any step independently
/cpm:spec I need a REST API for inventory management
/cpm:do 3  # work on a specific task

Key Features:

• Party mode — named agent personas discuss, debate, and disagree constructively
• Consult mode — focused one-to-one expert dialogue with invite/dismiss and lead transfer
• Multi-perspective insights woven into discover and spec phases
• Product ideation — explore vision, value propositions, and user journeys before requirements
• Architecture exploration — facilitated ADRs with trade-off analysis and dependency mapping
• Facilitated conversations, not forms — builds on your answers
• One topic at a time with user-gated progression
• Scales depth to complexity — skips phases that don't add value
• MoSCoW prioritisation for requirements
• Architecture decisions with rationale and alternatives (references existing ADRs)
• Spec requirement traceability — stories link back to the requirements they satisfy
• Right-sized epics and stories with acceptance criteria and dependencies
• Testing thread through the pipeline — spec defines test approach tags, epics propagate them to criteria and generate testing tasks, do discovers and runs tests in verification gates
• Task execution loop with acceptance criteria verification and ADR awareness
• Test runner discovery — convention-based detection from project config files, cached per session
• Epic-level verification — completed epics are checked against their source spec
• Spec coverage roll-up — one command joins every epic's coverage matrix back to the spec's requirements and answers "is this spec fully delivered?", naming untraced requirements first
• Autonomous spec delivery — point /cpm:ralph at a spec and it generates the epics, then works them; the loop stops on the roll-up script's verdict, not on its own judgement
• Spec, ADR, and test coverage compliance review dimensions
• Lightweight retros with testing gap analysis that feed forward into the next planning cycle
• Adversarial review — agent personas challenge assumptions, spot gaps, and flag risks with optional autofix
• Independent codebase audit — nine dimensions of code health with file:line citations, prioritised findings, and handoff to spec/library/quick
• Course correction — surgically amend any artefact with cascading downstream updates (5 artifact types)
• Audience-aware artifact transformation — present planning artifacts to any audience in any format
• Two-tier template system — structural (fixed data contracts) and presentational (overridable)
• Project reference library — import docs that skills auto-discover and use as context
• Archive — clean up completed artefacts with staleness heuristics and chain detection
• Project status reconnaissance — scan artifacts and git history, produce a narrative briefing with next steps
• Customisable agent roster — override default personas per project
• Compaction resilience — seamlessly survives Claude Code context compaction, with /cpm:clean for on-demand cleanup of leftover session-state files

Companion tool — cpm board: a standalone terminal UI (cpm/tools/board/) that shows the CPM status of every project you register — a three-column Projects → Epics → Stories browser — and launches the right /cpm:* session for each without leaving the board. It reads each project's docs/ planning artifacts read-only. See the board README.

/cpm:ralph needs a Stop hook, and CPM does not ship one. The autonomous loop is driven by a Stop hook from a separate plugin; every other skill works without one. Install:

/plugin marketplace add ninthspace/ralph-loop
/plugin install ralph-loop@ninthspace-ralph

ralph-loop@ninthspace-ralph 1.2.0 or later is the supported configuration — it is what CPM's documented loop behaviour is written against. The loop also runs on ralph-loop@claude-plugins-official and ralph-wiggum@claude-code-plugins, and /cpm:ralph probes whichever hook is installed rather than checking a name or a version. What you lose below 1.2.0 is not an error message: an unreadable transcript ends the run silently and looks like a clean finish, every <promise> in the transcript is the hook's own reminder rather than the model's, and active: false in the loop's state file does nothing at all. See the fork's README for what each change does.

View full documentation | Interactive Training Guide

JS/TS Simplifier (v1.0.0)

Simplify and improve JavaScript and TypeScript code across an entire codebase

A skill that scans all JS/TS files (or a configurable subset) and applies clarity, consistency, and maintainability improvements while preserving exact functionality. Unlike targeted simplification of recently changed files, this skill works across the whole codebase.

Three parallel analysis agents:

• Modern Syntax — ES2015+ and ES2020+ upgrades (optional chaining, nullish coalescing, async/await, const/let)
• Code Quality — Dead code removal, conditional simplification, naming improvements, error handling
• Structure & Reuse — DRY violations, module organisation, function complexity, async patterns

Quick Start:

# Simplify all JS/TS files in the project
/js-simplify

# Narrow to a specific directory
/js-simplify src/

# Only git-modified files
/js-simplify only changed

# Focus on a specific pattern
/js-simplify focus on async patterns

Key Features:

• Parallel three-agent analysis for comprehensive coverage
• Respects project conventions (CLAUDE.md, ESLint, Prettier, tsconfig)
• Configurable scope — all files, specific directories, globs, or git-changed only
• Safety-first — never changes what the code does, only how it does it
• Flags ambiguous cases for manual review rather than auto-applying

Supported File Types:

• .js, .mjs, .cjs, .jsx, .ts, .tsx

View full documentation

Filament Mockup (v1.1.0)

Build high-fidelity Filament v5 admin mockups for stakeholder sign-off

A skill that turns a product brief or spec into a single self-contained HTML file that looks pixel-accurate to a real Filament v5 admin panel — clickable enough to walk a stakeholder through every screen and flow, and throwaway by design (the real Filament build regenerates all of it natively). Mockups use the real captured Filament theme CSS and Filament's exact fi-* markup, so what stakeholders sign off on is what gets built. Not for production Filament code or customer-facing/front-end mockups.

Workflow:

• Capture — lift the compiled theme CSS and design tokens from a real Filament v5 panel
• Inventory — build an FR → screen matrix so every element traces back to a numbered functional requirement
• Build — reuse Filament's exact fi-* grammar; mark genuinely custom components with the mk- namespace
• Verify — measure with Playwright rather than eyeballing, then sign off with a coverage audit
• Hand off — write the durable routing table (docs/mockups/surface-routing.md) so the downstream builder mockup-to-filament knows which surfaces it owns (works stand-alone — no brief-to-mockups prerequisite)

Quick Start:

# Turn a brief/spec into clickable admin screens
create a Filament mockup from docs/specifications/05-spec-admin-panel.md

# Or describe it directly
mock up the admin panel for this PRD

Key Features:

• Single self-contained HTML file — opens by file://, zero environment to stand up
• Maximum fidelity to Filament's real design language (captured theme, Albert Sans, standard layouts)
• Every element traces to a functional requirement — invented UI is flagged, not silently added
• Visible mk- vs fi-* boundary distinguishes mockup scaffolding from real Filament
• Bundled scaffold, capture/verify scripts, and an fi-* grammar cheat-sheet
• Part of the mockup→build family — a producer whose output the builders consume (mockup-to-filament for Filament, mockup-to-blade for bespoke); emits a routing handoff naming the lane per surface

Requires: Node + Playwright for the capture/verify scripts (npm i -D playwright && npx playwright install chromium).

View full documentation

Removing Plugins (when in Claude Code)

# Uninstall individual plugins
/plugin uninstall noteplan@claude-code-marketplace
/plugin uninstall php-lsp@claude-code-marketplace
/plugin uninstall cpm@claude-code-marketplace
/plugin uninstall js-simplifier@claude-code-marketplace
/plugin uninstall filament-mockup@claude-code-marketplace

# Remove the entire marketplace
/plugin marketplace remove ninthspace-marketplace

License

MIT - See LICENSE for details

Contributing

Contributions welcome! Please:

1. Follow the existing plugin structure
2. Include comprehensive documentation
3. Add tests where applicable
4. Update the marketplace manifest
5. Submit a pull request

Support

For issues or questions:

• Open an issue on GitHub
• Check plugin-specific documentation
• Review the Claude Code plugin docs

Changelog

See individual plugin CHANGELOG.md files for version history.

Author

Chris Aves

Links

• Plugin Documentation
• Marketplace Documentation
• Claude Code Documentation