CommandVault MCP

Search CommandVault's library of AI commands (prompts, CLAUDE.md/AGENTS.md templates, Cursor rules, slash commands, subagents). Returns slugs to pass to get_command.

{
  "type": "object",
  "properties": {
    "tag": {
      "type": "string",
      "description": "Tag, e.g. 'claude-code', 'cursor', 'nextjs'."
    },
    "kind": {
      "enum": [
        "prompt",
        "workflow",
        "template",
        "snippet",
        "checklist",
        "skill",
        "mcp_server"
      ],
      "type": "string",
      "description": "Only this kind. 'skill' = SKILL.md Agent Skills, 'mcp_server' = MCP servers, 'template' = files like CLAUDE.md or Cursor rules."
    },
    "sort": {
      "enum": [
        "top",
        "hot",
        "new"
      ],
      "type": "string",
      "default": "top"
    },
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 25,
      "minimum": 1
    },
    "query": {
      "type": "string",
      "description": "Words to look for in titles, summaries and text. Leave empty to browse."
    },
    "community": {
      "type": "string",
      "description": "Community slug, e.g. 'agents', 'frontend', 'marketing'. See list_communities."
    }
  }
}

{
  "title": "Search commands",
  "readOnlyHint": true,
  "openWorldHint": false
}

Fetch the full text of a command by slug, with {{variables}} filled from the values you pass. Unfilled variables are listed so you can ask the user for them.

{
  "type": "object",
  "required": [
    "slug"
  ],
  "properties": {
    "slug": {
      "type": "string",
      "description": "The command's slug from search_commands or its URL (commandvault.ai/commands/<slug>)."
    },
    "variables": {
      "type": "object",
      "description": "Values for the command's {{blanks}}, keyed by variable name.",
      "additionalProperties": {
        "type": "string"
      }
    }
  }
}

{
  "title": "Get a command",
  "readOnlyHint": true,
  "openWorldHint": false
}

All CommandVault communities with their slug and number of commands.

{
  "type": "object",
  "properties": {}
}

{
  "title": "List communities",
  "readOnlyHint": true,
  "openWorldHint": false
}

Curated collections (starter kits such as 'claude-code-starter-kit' or 'cursor-rules-by-framework'). Pass a slug to get the commands in one.

{
  "type": "object",
  "properties": {
    "slug": {
      "type": "string",
      "description": "Optional collection slug to list its commands."
    }
  }
}

{
  "title": "List collections",
  "readOnlyHint": true,
  "openWorldHint": false
}

The commands people are copying, running and voting for most this week.

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 25,
      "minimum": 1
    }
  }
}

{
  "title": "Trending commands",
  "readOnlyHint": true,
  "openWorldHint": false
}

Advertised by the server during the last successful anonymous check.

This commands ensures that when a user lands on any page of an application, the user gets taken to the very top of that page, and not the position the user was on the previous page.

goal

Exceptions, e.g. 'back/forward navigation and in-page #anchors', or 'none' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Keep users oriented on bad URLs

goal

The page or action that helps them most, e.g. 'the dashboard and search' or 'the product catalog' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Persist a user’s light/dark preference across sessions and devices.

goal

e.g. 'follow the system setting' or 'always light first' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Verify new accounts seamlessly

goal

e.g. 'nothing until verified' or 'browse but not post' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

A full template including initial command and knowledge base to build a local business directory service that you can apply to your local community and start helping small businesses while making a profit.

goal

e.g. 'home cleaning services in St. Augustine, FL' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Reduce server load while users type

goal

e.g. 'products by name and SKU' or 'help articles' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Prevent data loss on long forms

goal

e.g. 'the job application form' or 'the long blog post editor' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Give Claude Code your Django project's commands, app layout and rules in one file.

project_name

e.g. clinic-portal (required)

what_it_does

One line, e.g. patient booking and records for small clinics (required)

python_version

e.g. 3.12 (required)

django_version

e.g. 5.2 (required)

database

e.g. PostgreSQL 16 (required)

project_package

The folder with settings.py, e.g. config (required)

test_command

The exact command you run, e.g. pytest -q or python manage.py test

Advertised by the server during the last successful anonymous check.

Save as .claude/commands/plan.md to make Claude investigate and plan before writing any code.

project_notes

Project notes (required)

Advertised by the server during the last successful anonymous check.

Turn notes into spaced-repetition cards that test understanding.

notes

Notes (required)

Advertised by the server during the last successful anonymous check.

Smooth perceived performance

goal

e.g. 'the dashboard cards and the orders table' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Fast, secure password recovery

goal

e.g. 'the dashboard, already logged in' or 'the login page' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

Limit sensitive features to the right people

goal

e.g. 'admin: everything; editor: create and edit posts; viewer: read only' (required)

stack

Framework, language and versions, e.g. Next.js 16 App Router + TypeScript

Advertised by the server during the last successful anonymous check.

A ready-to-edit CLAUDE.md that tells Claude Code how your Next.js app is built, tested and deployed.

app_name

App name (required)

one_line_description

One line description (required)

who_uses_it

Who uses it (required)

package_manager

Package manager (required)

test_command

Test command (required)

styling

Styling (required)

hosting

Hosting (required)

Advertised by the server during the last successful anonymous check.

Search Cloudflare's documentation (Workers, Pages, R2, D1, DNS and more) from your agent. No sign-in.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Turns a rough idea into an agreed design through focused questions before any code is written.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Project memory for Claude Code in a Rails codebase: commands, structure and Rails-specific rules.

project_name

e.g. shop-admin (required)

what_it_does

One line, e.g. inventory and orders for a small store (required)

ruby_version

e.g. 3.3 (required)

rails_version

e.g. 8.0 (required)

database

e.g. PostgreSQL 16 (required)

frontend

e.g. Hotwire (Turbo + Stimulus) or React via jsbundling (required)

job_backend

e.g. Solid Queue or Sidekiq (required)

test_command

The exact command you run, e.g. bin/rails test or bundle exec rspec

Advertised by the server during the last successful anonymous check.

Search Vercel docs, and inspect your projects, deployments and build logs from your agent.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Up-to-date guidance for building apps on the Claude API and Anthropic SDKs.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A .claude/agents/test-writer.md subagent that writes focused tests in your project's style.

project_name

Project name (required)

test_framework

Test framework (required)

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

A reusable AGENTS.md that stops your AI agent from guessing.

project_description

Project description (required)

stack

Stack (required)

Advertised by the server during the last successful anonymous check.

Tell Claude Code how your Go service is built, tested and structured.

project_name

e.g. ledger-svc (required)

what_it_does

One line, e.g. records payments and exposes a REST API (required)

go_version

e.g. 1.24 (required)

router

e.g. chi, gin or net/http (required)

database

e.g. PostgreSQL via pgx (required)

service

Name of the folder under cmd/, e.g. ledger (required)

Advertised by the server during the last successful anonymous check.

A .claude/agents/refactorer.md that simplifies code without changing behavior.

project_name

Project name (required)

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

A .claude/agents/code-reviewer.md that reviews changes for bugs, risk and simplicity.

project_name

Project name (required)

base_branch

Base branch (required)

Advertised by the server during the last successful anonymous check.

A .claude/agents/debugger.md that finds root causes instead of patching symptoms.

project_name

Project name (required)

Advertised by the server during the last successful anonymous check.

A .claude/agents/security-auditor.md that checks changes for the vulnerabilities attackers try first.

project_name

Project name (required)

audit_command

Audit command (required)

Advertised by the server during the last successful anonymous check.

Executes a plan by sending each task to a fresh subagent, with a review after every step.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Sends finished work to a reviewer subagent that checks it against the plan and requirements.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Save as .claude/commands/fix-issue.md: give it a GitHub issue number and it reproduces, fixes and tests.

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

Project memory for Claude Code in Laravel: Artisan commands, structure and conventions.

project_name

e.g. booking-app (required)

what_it_does

One line, e.g. appointment booking for salons (required)

php_version

e.g. 8.3 (required)

laravel_version

e.g. 12 (required)

database

e.g. MySQL 8 (required)

frontend

e.g. Livewire, Inertia + Vue, or Blade (required)

dev_command

The exact command you run, e.g. composer run dev or php artisan serve

Advertised by the server during the last successful anonymous check.

Project memory for Claude Code on a Python API: commands, layout, testing and the rules that keep it safe.

service_name

What the team calls this service, e.g. billing-api (required)

what_it_does

One line, e.g. issues invoices and handles Stripe webhooks (required)

python_version

e.g. 3.12 (required)

database

e.g. PostgreSQL 16 via SQLAlchemy (required)

setup_command

Exact command to install dependencies, e.g. uv sync (required)

module

Where the FastAPI app object lives, e.g. app.main (for uvicorn app.main:app) (required)

package

Top-level Python package for type checks, e.g. app (required)

Advertised by the server during the last successful anonymous check.

Search CommandVault's prompts, CLAUDE.md templates, Cursor rules, skills and MCP servers from your editor.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Audits your dependency tree for advisories, abandoned packages and risky install scripts.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

One root AGENTS.md plus per-package overrides so Codex, Cursor, Jules and other agents work the same way everywhere.

monorepo_tool

e.g. pnpm workspaces + Turborepo, Nx, or npm workspaces (required)

web_app

Folder name under apps/, e.g. web (required)

web_owner

Team or person, e.g. @frontend (required)

api_app

Folder name under apps/, e.g. api (required)

api_owner

Team or person, e.g. @platform (required)

ui_owner

Team or person, e.g. @design-systems (required)

test_one_package

Exact command, e.g. pnpm --filter <package> test, npx nx test <package>, or turbo run test --filter=<package> (required)

check_changed

Exact command that lints, tests and type-checks what changed, e.g. turbo run lint test typecheck --filter=...[origin/main] or npx nx affected -t lint test typecheck (required)

Advertised by the server during the last successful anonymous check.

Describe your tables and ask in English; get correct, readable SQL.

database

Database (required)

schema

Schema (required)

question

Question (required)

Advertised by the server during the last successful anonymous check.

Read, create, merge, split, fill and OCR PDF files from inside Claude.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Design, test and tune your own Agent Skills, including evals and description tuning.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Push UI work past generic defaults toward a distinctive, deliberate visual design.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Search, read and write Notion pages and databases from your AI tool.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Create original posters and static art as PNG or PDF, guided by a design philosophy.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Tighter sentences, same personality.

text

Text (required)

Advertised by the server during the last successful anonymous check.

A first-session prompt that has Claude Code explore your repo and create CLAUDE.md, commands and settings.

safe_commands

Safe commands (required)

Advertised by the server during the last successful anonymous check.

Build Gradio web UIs and demos in Python, from simple forms to chatbots.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Run and troubleshoot Wrangler for local dev, previews, deploys and resources.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Step-by-step help upgrading Stripe API versions, SDKs, Stripe.js and webhooks.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Writes and rewrites landing, pricing and product page copy that sounds human and converts.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

How to write Claude Code hooks that validate, block or automate tool use.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A .cursor/rules/prisma.mdc for schema changes, migrations and efficient queries.

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

Write and run Terraform's native .tftest.hcl tests, including mocks and assertions.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Build and modify dbt models the right way: ref/source, tests and checking results with dbt show.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Query Cloudflare Workers logs and analytics to debug errors and performance from your agent.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Look up and manage PagerDuty incidents, services, schedules and on-call from a hosted server.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Search Shopify developer docs and introspect the Admin GraphQL schema to build Shopify apps and themes accurately.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Read and update monday.com boards, items, updates and docs from your agent.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Manage Todoist tasks, projects, sections, labels and comments from your agent.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A structured scratchpad tool that helps an agent break problems into revisable, branching steps.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Ask questions about any public GitHub repository and get answers from its generated wiki. No sign-in.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Search and fetch official Microsoft documentation (Azure, .NET, Microsoft 365, Windows) from your agent. No sign-in.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Pulls current, version-specific docs and code examples for thousands of libraries into your prompt, so the agent stops inventing old APIs.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Manage Supabase projects from your agent: run SQL, apply migrations, read logs and advisors, generate TypeScript types.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Search models, datasets, Spaces and papers on the Hugging Face Hub, and call Gradio Spaces as tools.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Learn a concept by explaining it simply and finding the gaps.

concept

Concept (required)

Advertised by the server during the last successful anonymous check.

Find, create and update Linear issues, projects and comments without leaving your editor.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Work with the Stripe API and search Stripe's docs and knowledge base from your agent.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Fast web search and page fetching built for AI agents, with clean text results.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Gives your assistant a persistent memory as a local knowledge graph of entities, relations and observations.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Fetches a web page and converts it to Markdown for the model to read.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Endpoints, payloads, status codes and edge cases before you write code.

feature

Feature (required)

stack

Stack (required)

Advertised by the server during the last successful anonymous check.

Create and edit Word .docx files with real formatting, tracked changes and comments.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Build, edit and QA PowerPoint presentations, from pitch decks to template updates.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Create, clean and analyze Excel and CSV files with working formulas and formatting.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A .cursor/rules/python.mdc for typed, tested, well-structured Python.

python_version

Python version (required)

dependency_tool

Dependency tool (required)

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

Build well-designed MCP servers in Python or TypeScript that wrap an external API.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Drive a local web app in a real browser to test flows, grab screenshots and read logs.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Build multi-component React + Tailwind + shadcn/ui apps bundled into a single HTML file.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Finds exploitable flaws in GitHub Actions workflows, like pwn requests and script injection.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A structured workflow for writing specs, proposals and decision docs together with Claude.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Patterns for flexible React components: compound components, context and fewer boolean props.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Converts a spec into a step-by-step implementation plan with small, testable tasks.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Enforces red-green-refactor: write a failing test first, then the minimal code to pass it.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A four-phase process that finds the root cause of a bug before any fix is attempted.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Search and read files in Dropbox through Dropbox's hosted OAuth server.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Sets up an isolated git worktree with a clean, passing baseline before feature work starts.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Requires Claude to run real checks and show evidence before saying work is finished.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Vercel's ranked rules for faster React and Next.js code: waterfalls, bundles, rendering.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Audits UI code against Vercel's web interface guidelines and reports issues by file and line.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Deploys the current project to Vercel and returns a preview or production URL.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Runs a planned, approved Semgrep scan of a codebase and merges results into SARIF.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Builds a CodeQL database and runs data-flow and taint analysis to find vulnerabilities.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Security-focused review of a PR or diff, with blast radius, history and test coverage checks.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Clear explanations with analogies and no jargon.

topic

Topic (required)

audience

Audience (required)

length

Length (required)

Advertised by the server during the last successful anonymous check.

Write and review property-based tests with Hypothesis, fast-check, proptest and more.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

A .cursor/rules/go.mdc for idiomatic errors, contexts, interfaces and tests.

test_command

Test command (required)

Advertised by the server during the last successful anonymous check.

Use the hf CLI to download, upload and manage models, datasets, Spaces and jobs on the Hub.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Fine-tune language and vision models with TRL or Unsloth on Hugging Face cloud GPUs.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Reviews the changes on your branch for bugs, security holes and quality issues.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Loops on a pull request, fixing CI failures and review feedback until checks pass.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Helps pick and use the right Cloudflare products for apps, APIs, agents and storage.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Build stateful AI agents on Cloudflare with the agents package.

No arguments advertised.

Advertised by the server during the last successful anonymous check.

Official guidance for any Supabase work: database, auth, RLS, storage, edge functions, CLI.

No arguments advertised.