@mohammadhprp/system-prompt 0.10.1

package/framework/mcps/laravel-boost/install.md

@@ -0,0 +1,106 @@
+# Laravel Boost Installation
+
+## Requirements
+
+- Laravel application root as the working directory.
+- PHP available on `PATH`.
+- Composer available on `PATH`.
+- An MCP-compatible AI client.
+
+## Installation
+
+Run these commands from the Laravel application root:
+
+```bash
+composer require laravel/boost --dev
+php artisan boost:install
+```
+
+During `boost:install`, select the AI agents and features you want Boost to configure.
+
+## Configuration
+
+Boost can configure supported agents during installation. If manual registration is needed, use the official server details:
+
+| Field | Value |
+| --- | --- |
+| Server name | `laravel-boost` |
+| Command | `php` |
+| Args | `artisan`, `boost:mcp` |
+
+### Config file examples
+
+Use the examples in [`configs/`](./configs/) as client-specific starting points:
+
+| Client | Example | Typical location |
+| --- | --- | --- |
+| OpenCode | [`configs/opencode.json`](./configs/opencode.json) | OpenCode MCP configuration. |
+
+Because client configuration paths change over time, prefer the client's official documentation for the exact file location. Keep the command and args unchanged unless your Laravel app runs in a wrapper environment such as Docker, Sail, DDEV, or a remote container.
+
+## Client Setup Notes
+
+### OpenCode
+
+Copy [`configs/opencode.json`](./configs/opencode.json) into your OpenCode project root or user config directory and restart the OpenCode agent. The server will start automatically when the agent needs it.
+
+## Verification
+
+Run the MCP server command directly from the Laravel application root:
+
+```bash
+php artisan boost:mcp
+```
+
+Then verify from your AI client:
+
+- The `laravel-boost` MCP server is listed.
+- The server starts without errors.
+- Boost tools such as application info, database schema, logs, or documentation search are visible to the agent.
+
+## Updating
+
+Update Boost resources after package updates or when generated agent files need to be refreshed:
+
+```bash
+php artisan boost:update
+```
+
+To discover newly installed package guidelines and skills:
+
+```bash
+php artisan boost:update --discover
+```
+
+Optional Composer automation:
+
+```json
+{
+  "scripts": {
+    "post-update-cmd": [
+      "@php artisan boost:update --ansi"
+    ]
+  }
+}
+```
+
+## Uninstalling
+
+1. Remove the package:
+
+```bash
+composer remove laravel/boost --dev
+```
+
+2. Remove the `laravel-boost` MCP entry from AI client configuration files.
+3. Remove generated Boost files if your project does not keep them:
+   - `.mcp.json`
+   - `boost.json`
+   - generated agent guideline files such as `CLAUDE.md`, `AGENTS.md`, or client-specific directories
+
+## Common Issues
+
+- **Server does not start**: run `php artisan boost:mcp` manually and fix the first PHP, Composer, or Laravel error shown.
+- **Client cannot find the server**: confirm the config is in the location used by that client and the server name is `laravel-boost`.
+- **Wrong working directory**: run the AI client from the Laravel application root or configure the client to start the server there.
+- **Wrapper environment required**: replace `php artisan boost:mcp` with the command required by your local environment only when necessary.

package/framework/mcps/laravel-boost/troubleshooting.md

@@ -0,0 +1,119 @@
+# Laravel Boost Troubleshooting
+
+## Installation failure
+
+**Problem**
+
+`composer require laravel/boost --dev` fails.
+
+**Cause**
+
+Composer cannot resolve dependencies, PHP is missing, PHP extensions are missing, or the command is not being run from a Laravel application root.
+
+**Solution**
+
+- Run `php -v` and `composer --version`.
+- Run the command from the Laravel application root.
+- Resolve Composer dependency conflicts shown in the error output.
+- Install missing PHP extensions required by the Laravel application.
+
+## Server won't start
+
+**Problem**
+
+The AI client reports that `laravel-boost` failed to start.
+
+**Cause**
+
+The client cannot run `php artisan boost:mcp`, the working directory is wrong, dependencies are not installed, or the Laravel app fails during bootstrap.
+
+**Solution**
+
+- Run `php artisan boost:mcp` manually from the Laravel application root.
+- Fix the first Laravel, PHP, Composer, or environment error shown.
+- Ensure `vendor/` exists by running `composer install`.
+- Configure the client to start the server from the Laravel application root.
+
+## Configuration errors
+
+**Problem**
+
+The MCP server is not listed, or the client ignores the config file.
+
+**Cause**
+
+The config file is in the wrong location, the JSON is invalid, or the client expects a different configuration wrapper.
+
+**Solution**
+
+- Validate the JSON with `python -m json.tool <file>` or another JSON parser.
+- Confirm the client-specific config location in the client's documentation.
+- Keep the server name as `laravel-boost`.
+- Use the official command `php` and args `artisan`, `boost:mcp` unless your local environment requires a wrapper.
+
+## Authentication issues
+
+**Problem**
+
+Boost tools cannot access application features that depend on authenticated users or external services.
+
+**Cause**
+
+MCP tools run in the local Laravel application context and do not automatically create browser sessions, API tokens, or third-party service credentials.
+
+**Solution**
+
+- Use local test accounts and seeded data when possible.
+- Provide required `.env` values for local-only services.
+- Avoid exposing production secrets to AI clients.
+- Prefer tests or factories for authenticated workflows.
+
+## Permission problems
+
+**Problem**
+
+Boost cannot read logs, write generated files, or access project paths.
+
+**Cause**
+
+Filesystem ownership, container mounts, or restricted directories prevent PHP or the AI client process from accessing required files.
+
+**Solution**
+
+- Check ownership and permissions for `storage/`, `bootstrap/cache/`, generated agent files, and client config files.
+- Run the AI client and PHP process as a user with access to the project.
+- In containerized environments, ensure the project is mounted and writable where needed.
+
+## Version mismatch
+
+**Problem**
+
+Boost behavior or generated files do not match the documentation or the selected AI client.
+
+**Cause**
+
+The installed `laravel/boost` version, Laravel version, AI client version, or MCP support differs from the documented version.
+
+**Solution**
+
+- Check installed versions with Composer and the AI client's version command.
+- Run `composer update laravel/boost --dev` when appropriate.
+- Run `php artisan boost:update --discover` after dependency changes.
+- Re-run `php artisan boost:install` if client support files need to be regenerated.
+
+## Database query risk
+
+**Problem**
+
+An agent proposes running write queries or destructive Tinker code through Boost.
+
+**Cause**
+
+Boost exposes tools that can execute code or queries in the Laravel application context.
+
+**Solution**
+
+- Use a local or disposable database for agent work.
+- Require human approval before write queries, migrations, seeders, or destructive commands.
+- Ask the agent to show the exact query or code before execution.
+- Back up important data before allowing inspection against non-disposable databases.

package/framework/mcps/notion-mcp/README.md

@@ -0,0 +1,58 @@
+# Notion MCP
+
+## Overview
+
+Notion MCP is a hosted, officially maintained MCP server by Notion that lets AI agents search, read, and write content in a Notion workspace. It uses OAuth authentication and communicates over Streamable HTTP — no API tokens or local infrastructure required.
+
+Official source:
+
+- [Notion MCP get-started guide](https://developers.notion.com/guides/mcp/get-started-with-mcp)
+- [Tool reference](https://developers.notion.com/guides/mcp/mcp-supported-tools)
+- [GitHub: makenotion/notion-mcp-server](https://github.com/makenotion/notion-mcp-server) (open-source server, no longer actively maintained)
+
+## Features
+
+- Search across Notion workspace and connected tools (Slack, Google Drive, Jira).
+- Fetch page and database content by URL or ID.
+- Create, update, move, and duplicate pages.
+- Create and update databases, data sources, and views (table, board, list, calendar, timeline, gallery, form, chart, map, dashboard).
+- Query databases with SQL or using pre-defined views.
+- Query meeting notes for the current user.
+- Add and list comments and discussions.
+- Get workspace teams and users.
+- Async support for long-running create and update operations.
+- OAuth authentication — no manual token management.
+
+## Supported AI Clients
+
+Notion MCP officially documents setup for Claude Code, Cursor, VS Code (GitHub Copilot), Claude Desktop, Windsurf, ChatGPT, Codex, and Antigravity.
+
+- OpenCode
+
+## When to Use
+
+Use Notion MCP when an AI coding agent needs to read or write content in a Notion workspace.
+
+Good fits:
+
+- Creating project documentation pages from code context.
+- Searching for existing notes, specs, or meeting notes.
+- Updating task statuses, tracking work, or filing bugs.
+- Querying databases for structured data.
+- Managing comments and discussions on pages.
+- Fetching user and team information from the workspace.
+
+Avoid using it for fully automated or headless workflows — Notion MCP requires interactive OAuth authentication.
+
+## Requirements
+
+- A Notion workspace with appropriate permissions.
+- An MCP-compatible AI client that supports remote HTTP (Streamable HTTP or SSE) MCP servers.
+- For some tools (search, query, meeting notes): a Business plan or higher with Notion AI.
+
+## Related Skills
+
+Relevant skills in this repository:
+
+- [`backend-engineer`](../../skills/backend-engineer/SKILL.md): creating and updating project documentation in Notion.
+- [`backend-best-practices`](../../skills/backend-best-practices/SKILL.md): backend best practices for API design, testing, and debugging.

package/framework/mcps/notion-mcp/capabilities.md

@@ -0,0 +1,118 @@
+# Notion MCP Capabilities
+
+## What It Can Do
+
+18 tools for searching, reading, and writing content in a Notion workspace.
+
+### Search & Fetch
+
+| Tool | Description |
+| --- | --- |
+| `notion-search` | Search across workspace and connected tools (Slack, Drive, Jira); requires Notion AI for cross-tool search |
+| `notion-fetch` | Fetch content from a page, database, or data source by URL or ID; pass `self` to get workspace and user identity |
+
+### Create
+
+| Tool | Description |
+| --- | --- |
+| `notion-create-pages` | Create one or more pages with properties, content, icons, and covers; supports database templates and async mode |
+| `notion-create-database` | Create a new database with specified properties, initial data source, and initial view |
+| `notion-create-view` | Create a view on a database (table, board, list, calendar, timeline, gallery, form, chart, map, dashboard) |
+| `notion-create-comment` | Add a comment to a page, block-level content, or reply to an existing discussion |
+
+### Update
+
+| Tool | Description |
+| --- | --- |
+| `notion-update-page` | Update page properties, content, icon, or cover; supports database templates and async mode |
+| `notion-update-data-source` | Update a data source's properties, name, description |
+| `notion-update-view` | Update a view's name, filters, sorts, display configuration |
+
+### Move & Duplicate
+
+| Tool | Description |
+| --- | --- |
+| `notion-move-pages` | Move one or more pages or databases to a new parent |
+| `notion-duplicate-page` | Duplicate a page within the workspace (async) |
+
+### Query
+
+| Tool | Description |
+| --- | --- |
+| `notion-query-data-sources` | Query data sources with SQL across multiple databases; requires Business plan or higher with Notion AI |
+| `notion-query-database-view` | Query a database using a pre-defined view's filters and sorts; requires Business plan or higher with Notion AI |
+| `notion-query-meeting-notes` | Query the current user's meeting notes; requires Business plan or higher with Notion AI |
+
+### Comments
+
+| Tool | Description |
+| --- | --- |
+| `notion-get-comments` | List all comments and discussions on a page, including resolved threads |
+
+### Workspace
+
+| Tool | Description |
+| --- | --- |
+| `notion-get-teams` | List teams (teamspaces) in the current workspace |
+| `notion-get-users` | List workspace members and guests with search by name or email; pass `self` for current user |
+
+### Async Tasks
+
+| Tool | Description |
+| --- | --- |
+| `notion-get-async-task` | Check the status of an async task (duplicate, large create/update) |
+
+### OpenAI Client Note
+
+When connecting from an OpenAI-based client (ChatGPT), `notion-fetch` and `notion-search` appear without the `notion-` prefix as `fetch` and `search`.
+
+## What It Cannot Do
+
+- It cannot work without interactive OAuth authentication — no PAT or bearer token support.
+- It cannot upload images or file attachments (on the roadmap).
+- It cannot access Notion content outside the authenticated user's permissions.
+- Some tools require a Business plan or higher with Notion AI (search, query tools).
+- It cannot operate fully headless or automated — a human must complete the OAuth flow.
+- It cannot be self-hosted (use the deprecated open-source server for that).
+
+## Best Practices
+
+- Use `notion-fetch` with `self` to verify the workspace and user identity after connecting.
+- Use `notion-fetch` with a page URL to understand its structure before making changes.
+- Use `notion-search` first to find relevant pages, then `notion-fetch` to read details.
+- Use `allow_async: true` for large page create/update operations, then poll with `notion-get-async-task`.
+- Fetch a database first to see available templates before creating pages with them.
+- Use `notion-get-users` to look up user IDs before assigning or mentioning users.
+- Stay within rate limits: 180 requests/min average (shared across all tools), 30 requests/min for search.
+
+## Common Workflows
+
+### Search and read content
+
+1. Call `notion-search` to find pages matching a topic.
+2. Call `notion-fetch` with a page URL to read its full content.
+3. Call `notion-get-comments` to see discussions on the page.
+
+### Create a page from a template
+
+1. Call `notion-fetch` on a database to see available templates.
+2. Call `notion-create-pages` with the template reference and desired properties.
+3. Optionally set an icon and cover image.
+
+### Update a task status
+
+1. Call `notion-fetch` with the page URL to get the current properties.
+2. Call `notion-update-page` to change the status property.
+3. Call `notion-create-comment` to leave a note about the change.
+
+### Query a database
+
+1. Call `notion-fetch` on the database URL to see its schema and views.
+2. Call `notion-query-database-view` with a view name to get filtered results.
+3. Or call `notion-query-data-sources` with SQL for custom queries.
+
+### Duplicate and populate a template
+
+1. Call `notion-duplicate-page` on a template page.
+2. Poll with `notion-get-async-task` until complete.
+3. Call `notion-update-page` on the new copy to fill in properties.

package/framework/mcps/notion-mcp/configs/opencode.json

@@ -0,0 +1,9 @@
+{
+  "mcp": {
+    "notion": {
+      "type": "remote",
+      "enabled": true,
+      "url": "https://mcp.notion.com/mcp"
+    }
+  }
+}

package/framework/mcps/notion-mcp/install.md

@@ -0,0 +1,70 @@
+# Notion MCP Installation
+
+## Requirements
+
+- A Notion workspace with appropriate permissions.
+- An MCP-compatible AI client that supports remote HTTP MCP servers (Streamable HTTP or SSE).
+- No local infrastructure — the server is hosted by Notion at `https://mcp.notion.com/mcp`.
+
+## Configuration
+
+Notion MCP is a remote Streamable HTTP server — no local command or install needed. Authentication uses OAuth, completed interactively on first use.
+
+| Field | Value |
+| --- | --- |
+| Server name | `notion` |
+| Transport type | Streamable HTTP (recommended) |
+| URL | `https://mcp.notion.com/mcp` |
+
+### Config file examples
+
+Use the examples in [`configs/`](./configs/) as client-specific starting points:
+
+| Client | Example | Typical location |
+| --- | --- | --- |
+| OpenCode | [`configs/opencode.json`](./configs/opencode.json) | OpenCode MCP configuration. |
+
+### OpenCode
+
+Copy [`configs/opencode.json`](./configs/opencode.json) into your OpenCode project root or user config directory and restart the OpenCode agent. When you use a Notion tool for the first time, complete the OAuth flow to connect your workspace.
+
+### For clients that only support local stdio servers
+
+Some MCP clients only support local stdio servers. Use the [mcp-remote](https://www.npmjs.com/package/mcp-remote) bridge to connect:
+
+```json
+{
+  "mcpServers": {
+    "notion": {
+      "command": "npx",
+      "args": ["-y", "mcp-remote", "https://mcp.notion.com/mcp"]
+    }
+  }
+}
+```
+
+## Verification
+
+After configuring and restarting your AI client, use a Notion tool for the first time to trigger the OAuth flow.
+
+Recommended first prompt:
+
+```
+Fetch "self" to see which workspace and user this Notion connection is for.
+```
+
+## Updating
+
+Notion MCP is hosted by Notion — no update steps needed. The latest version is always available at `https://mcp.notion.com/mcp`.
+
+## Uninstalling
+
+Remove the `notion` MCP entry from AI client configuration files. Optionally, clear the OAuth connection in your Notion workspace: **Settings → Connections → Notion MCP**.
+
+## Common Issues
+
+- **Authentication fails**: make sure you complete the OAuth flow when prompted. Try disconnecting and reconnecting in the tool's MCP settings.
+- **Permission errors**: check that you have the correct permissions in the Notion workspace for the pages or databases you're trying to access.
+- **Client does not support remote servers**: use the `mcp-remote` bridge as a stdio wrapper to connect to the remote server.
+- **Tool not found**: some clients prefix tools with `notion-`. For OpenAI-based clients (ChatGPT), `notion-fetch` and `notion-search` may appear as `fetch` and `search`.
+- **Rate limited**: standard API limits apply (180 requests/min average, 30 requests/min for search). Prompt the agent to reduce parallel operations.

package/framework/mcps/notion-mcp/troubleshooting.md

@@ -0,0 +1,95 @@
+# Notion MCP Troubleshooting
+
+## Connection fails
+
+**Problem**
+
+The AI client reports that `notion` cannot connect.
+
+**Cause**
+
+The client does not support remote HTTP MCP servers, or the URL is misconfigured.
+
+**Solution**
+
+- Verify the URL is `https://mcp.notion.com/mcp` (Streamable HTTP) or `https://mcp.notion.com/sse` (SSE).
+- If your client only supports local stdio servers, use the `mcp-remote` bridge: `npx -y mcp-remote https://mcp.notion.com/mcp`.
+- Check the client's documentation for how to add a remote MCP server.
+
+## Authentication fails
+
+**Problem**
+
+The connection succeeds but tools return authentication errors.
+
+**Cause**
+
+The OAuth flow was not completed or the token has expired.
+
+**Solution**
+
+- Make sure you complete the OAuth flow when prompted by the client.
+- Try disconnecting and reconnecting. Look for a "Clear authentication" or "Disconnect" option in the client's MCP settings.
+- Re-add the MCP server to trigger a fresh OAuth flow.
+
+## Permission errors
+
+**Problem**
+
+Tools return 403 or "not found" errors.
+
+**Cause**
+
+The authenticated user does not have access to the requested page, database, or workspace.
+
+**Solution**
+
+- Verify you have the correct permissions in the Notion workspace.
+- Check that the page or database is shared with your user or team.
+- Use `notion-get-teams` to confirm which teams you can access.
+
+## Tool requires Notion AI
+
+**Problem**
+
+`notion-search`, `notion-query-data-sources`, `notion-query-database-view`, or `notion-query-meeting-notes` return an upgrade prompt.
+
+**Cause**
+
+These tools require a Business plan or higher with Notion AI.
+
+**Solution**
+
+- Upgrade your workspace plan to Business or Enterprise with Notion AI enabled.
+- Use `notion-fetch` and standard page reading as an alternative.
+
+## Rate limited
+
+**Problem**
+
+Tools return 429 Too Many Requests errors.
+
+**Cause**
+
+API rate limits exceeded: 180 requests/min (general), 30 requests/min (search).
+
+**Solution**
+
+- Prompt the agent to reduce the number of parallel tool calls.
+- Space out requests — the time between sequential agent steps often keeps usage under the limit.
+- Wait before retrying.
+
+## Tool names differ in OpenAI clients
+
+**Problem**
+
+`notion-fetch` or `notion-search` tools are missing in ChatGPT.
+
+**Cause**
+
+OpenAI clients strip the `notion-` prefix — tools appear as `fetch` and `search`.
+
+**Solution**
+
+- Use `fetch` and `search` as the tool names in prompts for ChatGPT.
+- Other tools retain the `notion-` prefix.

package/framework/mcps/playwright-mcp/README.md

@@ -0,0 +1,50 @@
+# Playwright MCP
+
+## Overview
+
+Playwright MCP lets coding agents automate and interact with web pages using [Playwright](https://playwright.dev). It uses structured accessibility snapshots instead of pixel-based input, so no vision model is needed. Published and maintained by Microsoft.
+
+Official source:
+
+- [GitHub: microsoft/playwright-mcp](https://github.com/microsoft/playwright-mcp)
+
+## Features
+
+- **Fast and lightweight**: Uses Playwright's accessibility tree, not pixel-based input or screenshots.
+- **LLM-friendly**: Operates purely on structured data — no vision models required.
+- **Cross-browser**: Supports Chromium, Firefox, and WebKit.
+- **Deterministic tool application**: Avoids ambiguity common with screenshot-based approaches.
+- **Persistent profiles**: Session data saved between runs with automatic workspace-scoped isolation.
+- **Headless support**: Run headed or headless.
+- **Capability system**: Opt-in tool categories (network, storage, devtools, vision, PDF).
+
+## Supported AI Clients
+
+Playwright MCP officially documents setup for Amp, Antigravity, Claude Code, Claude Desktop, Cline, Codex, Copilot, Cursor, Factory, Gemini CLI, Goose, Junie, Kiro, LM Studio, OpenCode, Qodo Gen, VS Code, Warp, and Windsurf.
+
+## When to Use
+
+Use Playwright MCP when an AI coding agent needs to navigate, interact with, or inspect web pages.
+
+Good fits:
+
+- Automating browser interactions (click, fill forms, navigate, type).
+- Taking accessibility snapshots for structured page understanding.
+- E2E test debugging and inspection.
+- Capturing screenshots and console logs.
+- Evaluating JavaScript in page context.
+- Simulating network conditions and mocking requests.
+- Managing cookies, localStorage, and session state.
+
+Avoid using it when no browser interaction is needed or when you need pixel-level visual inspection (use `--caps=vision` if coordinates are required).
+
+## Requirements
+
+- Node.js 18 or newer.
+- An MCP-compatible AI client.
+
+## Related Skills
+
+Relevant skills in this repository:
+
+- [`backend-best-practices`](../../skills/backend-best-practices/SKILL.md): backend best practices for testing, debugging, and performance analysis.