@mohammadhprp/system-prompt 0.12.1 → 0.12.3

package/framework/agents/backend-architect.md

@@ -1,146 +0,0 @@
----
-description: Master backend architect for designing scalable, secure, and maintainable server-side systems. Use for API design, database architecture, system architecture decisions, security implementation, performance optimization, and DevOps integration.
-mode: subagent
-temperature: 0.1
-permission:
-  bash: allow
-  read: allow
-  edit: deny
-  write: deny
-  patch: deny
-  grep: allow
-  glob: allow
-  list: allow
-  webfetch: allow
-  todoread: deny
-  todowrite: deny
----
-
-You are a master backend architect with deep expertise in designing scalable, secure, and maintainable server-side systems. You excel at making architectural decisions that balance immediate needs with long-term scalability.
-
-## Core Responsibilities
-
-### 1. API Design & Implementation
-- Design RESTful APIs following OpenAPI specifications
-- Implement GraphQL schemas when appropriate
-- Create proper versioning strategies
-- Implement comprehensive error handling
-- Design consistent response formats
-- Build proper authentication and authorization
-
-### 2. Database Architecture
-- Choose appropriate databases (SQL vs NoSQL)
-- Design normalized schemas with proper relationships
-- Implement efficient indexing strategies
-- Create data migration strategies
-- Handle concurrent access patterns
-- Implement caching layers (Redis, Memcached)
-
-### 3. System Architecture
-- Design microservices with clear boundaries
-- Implement message queues for async processing
-- Create event-driven architectures
-- Build fault-tolerant systems
-- Implement circuit breakers and retries
-- Design for horizontal scaling
-
-### 4. Security Implementation
-- Implement proper authentication (JWT, OAuth2)
-- Create role-based access control (RBAC)
-- Validate and sanitize all inputs
-- Implement rate limiting and DDoS protection
-- Encrypt sensitive data at rest and in transit
-- Follow OWASP security guidelines
-
-### 5. Performance Optimization
-- Implement efficient caching strategies
-- Optimize database queries and connections
-- Use connection pooling effectively
-- Implement lazy loading where appropriate
-- Monitor and optimize memory usage
-- Create performance benchmarks
-
-### 6. DevOps Integration
-- Create Dockerized applications
-- Implement health checks and monitoring
-- Set up proper logging and tracing
-- Create CI/CD-friendly architectures
-- Implement feature flags for safe deployments
-- Design for zero-downtime deployments
-
-## Technology Stack Expertise
-
-**Languages:** PHP, Python, Go, Rust
-**Frameworks:** Laravel, FastAPI, Gin
-**Databases:** PostgreSQL, MongoDB, Redis, MySQL
-**Message Queues:** RabbitMQ, Kafka, SQS
-**Cloud:** VPS, Vercel, Supabase
-
-## Architectural Patterns
-
-- Microservices with API Gateway
-- Event Sourcing and CQRS
-- Serverless with Lambda/Functions
-- Domain-Driven Design (DDD)
-- Hexagonal Architecture
-- Service Mesh with Istio
-
-## Workflow
-
-### Step 1: Understand Requirements
-Read conversation history, project structure, and existing code. Identify constraints: traffic patterns, data volume, latency requirements, team expertise, deployment environment. Load [`skills/backend-best-practices/SKILL.md`](../skills/backend-best-practices/SKILL.md), and referenced standards.
-
-### Step 2: Evaluate Tradeoffs
-For each architectural decision, consider:
-- **Complexity vs maintainability**: will this abstraction pay off in 3 months?
-- **Performance vs cost**: does the optimization justify the infrastructure expense?
-- **Speed vs correctness**: can we ship a safe subset now and iterate?
-- **Consistency vs availability**: which matters more for this specific operation?
-
-### Step 3: Design and Document
-Produce artifacts matching the task scope:
-- API contracts (OpenAPI or GraphQL schema)
-- Data model with entity relationships and indexes
-- System architecture diagram (text or component map)
-- Deployment architecture (containers, networking, scaling)
-- Security model (auth flows, RBAC matrix, encryption boundaries)
-
-### Step 4: Review Against Principles
-- Can each component change independently?
-- Is there a simpler architecture that satisfies the same requirements?
-- Are failure modes documented (timeout, retry, fallback, partial failure)?
-- Are operational concerns addressed (observability, deployability, rollback)?
-
-## Output Format
-
-```
-## Architecture Summary
-[High-level design and key decisions]
-
-## API Design
-[Contracts, versioning, error handling approach]
-
-## Data Model
-[Entity relationships, storage choices, caching]
-
-## System Architecture
-[Components, boundaries, communication patterns]
-
-## Security Model
-[Auth, authorization, input validation, encryption]
-
-## Performance Considerations
-[Caching, connection pooling, scaling strategy]
-
-## Tradeoffs & Risks
-[What was deferred, what could break, mitigation plan]
-```
-
-## Tone
-
-- **Pragmatic** — recommend what works for the actual constraints, not theoretical ideal
-- **Explicit about tradeoffs** — every decision has a cost; surface it
-- **Specific and actionable** — concrete schema snippets, config examples, code patterns
-- **Cost-aware** — consider infra cost, team velocity, and maintenance burden
-
-Return findings in response. Do not modify files.

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

@@ -1,51 +0,0 @@
-# GitHub MCP
-
-## Overview
-
-The GitHub MCP Server connects AI tools directly to GitHub. Browse code, manage issues and PRs, analyze commits, monitor Actions workflows, and automate development workflows through natural language.
-
-Official source:
-
-- [GitHub: github/github-mcp-server](https://github.com/github/github-mcp-server)
-
-## Features
-
-- Repository browsing — read files, search code, explore directory trees, analyze commits.
-- Issue & PR management — create, update, comment, review, and merge pull requests.
-- CI/CD intelligence — monitor workflow runs, analyze build failures, manage releases.
-- Code Analysis — examine security findings, review Dependabot alerts, understand code patterns.
-- Team collaboration — access discussions, manage notifications, analyze team activity.
-
-## Supported AI Clients
-
-GitHub MCP officially documents setup for Claude Code, VS Code, GitHub Copilot, Cursor, Windsurf, Zed, and OpenCode.
-
-- OpenCode
-
-## When to Use
-
-Use GitHub MCP when an AI coding agent needs direct access to GitHub.
-
-Good fits:
-
-- Browsing repository code, commits, and branches.
-- Creating, reviewing, and merging pull requests.
-- Managing issues, labels, and milestones.
-- Monitoring Actions workflows and debugging CI failures.
-- Analyzing Dependabot alerts and code scanning results.
-- Managing GitHub releases and gists.
-
-Avoid using it outside of GitHub-related workflows.
-
-## Requirements
-
-- A GitHub Personal Access Token with appropriate scopes (`repo` for private repos, `read:org` for org access).
-- An MCP-compatible AI client that supports remote HTTP MCP servers.
-
-## Related Skills
-
-Relevant skills in this repository:
-
-- [`code-review`](../../skills/code-review/SKILL.md): reviewing pull request diffs.
-- [`backend-best-practices`](../../skills/backend-best-practices/SKILL.md): backend best practices for testing, debugging, and security review.
-- [`pr`](../../commands/pr.md): preparing and submitting pull requests.

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

@@ -1,83 +0,0 @@
-# GitHub MCP Capabilities
-
-## What It Can Do
-
-The server provides toolsets across the following categories:
-
-### Context
-
-Tools that provide information about the authenticated user and organization membership.
-
-### Repositories
-
-Browse repository code, files, directory trees, commits, and branches. Search code across repositories.
-
-### Issues
-
-Create, read, update issues. Add comments, manage labels, assignees, and milestones.
-
-### Pull Requests
-
-Create, review, update, and merge pull requests. Add review comments, request changes, approve. List changed files and review diffs.
-
-### Actions
-
-List workflows, trigger workflow runs. Get workflow run details, job logs, and artifacts.
-
-### Code Security
-
-List and read code scanning alerts, Dependabot alerts.
-
-### Discussions
-
-Read and create discussions, manage comments.
-
-### Gists
-
-Create, read, update gists.
-
-### Users & Organizations
-
-Read user profiles, list teams, list organization members.
-
-### Additional Toolsets (remote server)
-
-- Copilot — assign Copilot to issues, request Copilot review.
-- Copilot Spaces — manage Copilot spaces.
-- GitHub Support Docs Search — answer GitHub product and support questions.
-
-## What It Cannot Do
-
-- It cannot operate without a valid GitHub token or OAuth session.
-- It cannot access repositories the token does not have permissions for.
-- It cannot directly execute code or deploy to production.
-- It does not provide real-time notifications or webhook management.
-- Some toolsets are only available on the remote server (Copilot, Support Docs Search).
-
-## Best Practices
-
-- Use a fine-grained PAT scoped to the minimum repositories and permissions needed.
-- Store the PAT in `.env` (under `.opencode/`) and never commit it to version control.
-- Use `GITHUB_TOOLSETS` environment variable (when running locally) to enable only the tool groups needed.
-- For read-only workflows, limit the PAT to read-only scopes.
-- Prefer the remote HTTP server for simplicity; use the local Docker server only when remote HTTP is not supported.
-
-## Common Workflows
-
-### Review a pull request
-
-1. Let the agent browse the PR diff and changed files.
-2. The agent can post review comments, approve, or request changes.
-3. Merge when ready.
-
-### Investigate a CI failure
-
-1. Ask the agent to list recent Actions workflow runs.
-2. The agent can read job logs to find the failure.
-3. Use the findings to propose a fix.
-
-### Manage issues
-
-1. List open issues with filtering by label, assignee, or state.
-2. Read issue details and comments.
-3. Create or update issues with labels, assignees, and milestones.

package/framework/mcps/github-mcp/configs/.env.example

@@ -1 +0,0 @@
-GITHUB_PERSONAL_ACCESS_TOKEN=

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

@@ -1,13 +0,0 @@
-{
-  "mcp": {
-    "github": {
-      "type": "remote",
-      "enabled": true,
-      "oauth": false,
-      "url": "https://api.githubcopilot.com/mcp/",
-      "headers": {
-        "Authorization": "Bearer {env:GITHUB_PERSONAL_ACCESS_TOKEN}"
-      }
-    }
-  }
-}

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

@@ -1,60 +0,0 @@
-# GitHub MCP Installation
-
-## Requirements
-
-- A GitHub account with API access.
-- A GitHub Personal Access Token (classic or fine-grained) with the following scopes:
-  - `repo` — full control of private repositories.
-  - `read:org` — read organization membership.
-- An MCP-compatible AI client that supports remote HTTP MCP servers.
-
-## Authentication
-
-This MCP uses the **remote GitHub MCP Server** hosted by GitHub at `https://api.githubcopilot.com/mcp/`. Authentication is handled via a `Bearer` token in the `Authorization` header.
-
-| Method | Best for |
-| --- | --- |
-| Personal Access Token (PAT) | Simplest local setup |
-| OAuth (via host support) | Better security when host supports it |
-
-## Configuration
-
-| Field | Value |
-| --- | --- |
-| Server name | `github` |
-| Type | `http` |
-| URL | `https://api.githubcopilot.com/mcp/` |
-| Auth header | `Authorization: Bearer {env:GITHUB_PERSONAL_ACCESS_TOKEN}` |
-
-### 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, ensure `GITHUB_PERSONAL_ACCESS_TOKEN` is set in your `.env` file, and restart the OpenCode agent.
-
-## Verification
-
-From your AI client:
-
-- The `github` MCP server is listed.
-- The server starts without errors.
-- Tools are visible to the agent.
-
-Recommended first prompt:
-
-```
-List my GitHub repositories and show me recent pull requests.
-```
-
-## Common Issues
-
-- **Server does not start**: confirm your client supports remote HTTP MCP servers.
-- **Authentication errors**: confirm the PAT has the required scopes and `GITHUB_PERSONAL_ACCESS_TOKEN` is set in your `.env` or environment.
-- **401 Unauthorized**: regenerate the token in GitHub **Settings → Developer settings → Personal access tokens**.
-- **Rate limiting**: GitHub API has rate limits; reduce the frequency of tool calls if you encounter 429 errors.

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

@@ -1,79 +0,0 @@
-# GitHub MCP Troubleshooting
-
-## Server won't start
-
-**Problem**
-
-The AI client reports that `github` failed to start.
-
-**Cause**
-
-The client does not support remote HTTP MCP servers, or the URL is misconfigured.
-
-**Solution**
-
-- Verify the `type` is `http` and `url` is `https://api.githubcopilot.com/mcp/`.
-- Check that your client supports remote MCP servers (VS Code 1.101+, Claude Desktop, Cursor, Windsurf, OpenCode).
-- If remote HTTP is not supported, use the local Docker version instead (see upstream docs).
-
-## Authentication errors
-
-**Problem**
-
-The server starts but returns 401 Unauthorized on tool calls.
-
-**Cause**
-
-The `GITHUB_PERSONAL_ACCESS_TOKEN` is missing, expired, or invalid.
-
-**Solution**
-
-- Verify `GITHUB_PERSONAL_ACCESS_TOKEN` is set: check your `.env` file or environment variables.
-- Confirm the token is a valid GitHub Personal Access Token with the required scopes.
-- Generate a new token in GitHub: **Settings → Developer settings → Personal access tokens → Fine-grained tokens**.
-- Ensure the token is not expired.
-
-## Rate limiting
-
-**Problem**
-
-Tools return 429 Too Many Requests errors.
-
-**Cause**
-
-GitHub API rate limits have been exceeded.
-
-**Solution**
-
-- Reduce the frequency of tool calls.
-- Wait for the rate limit window (typically 1 hour) to reset before retrying.
-- For higher limits, authenticate with a PAT that has the necessary scopes.
-
-## Token in version control
-
-**Problem**
-
-Risk of leaking the GitHub token to version control.
-
-**Solution**
-
-- Never hardcode the token in `opencode.json` or any config file.
-- Use `{env:GITHUB_PERSONAL_ACCESS_TOKEN}` in config files and store the actual value in `.env`.
-- Add `.env` to `.gitignore`.
-- Restrict `.env` file permissions: `chmod 600 .opencode/.env`.
-
-## Permission errors
-
-**Problem**
-
-Tools return 403 Forbidden or insufficient permissions.
-
-**Cause**
-
-The authenticated token does not have the required scopes or repository access.
-
-**Solution**
-
-- Verify the token has `repo` scope for private repositories.
-- For organization access, ensure `read:org` scope is included.
-- For fine-grained tokens, ensure the token has access to the target repository.

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

@@ -1,53 +0,0 @@
-# GitLab MCP
-
-## Overview
-
-GitLab MCP is a comprehensive GitLab MCP server for AI clients. Manage projects, merge requests, issues, pipelines, wiki, releases, tags, milestones, and more through stdio, SSE, and Streamable HTTP transports. Supports PAT, OAuth, read-only mode, dynamic API URLs, and remote authorization.
-
-Official source:
-
-- [GitHub: zereight/gitlab-mcp](https://github.com/zereight/gitlab-mcp)
-- [Documentation](https://zereight.github.io/gitlab-mcp/)
-
-## Features
-
-- Broad GitLab coverage — 117+ tools for projects, repository browsing, merge requests, issues, pipelines, wiki, releases, tags, labels, milestones, and more.
-- Flexible auth — Personal Access Token, local OAuth2 browser flow, MCP OAuth proxy, and per-request remote authorization.
-- Multiple transports — stdio for local clients, SSE for legacy clients, and Streamable HTTP for modern remote deployments.
-- Client-friendly setup — examples for Claude Code, Codex, OpenCode, Copilot, Cline, Roo Code, Cursor, and more.
-- Self-hosted ready — works with custom GitLab instances, proxy settings, and dynamic API URL routing.
-- Permission modes — `readonly`, `modify` (no delete), or `full`; tool filtering by toolset groups, allow-list, and deny-list.
-
-## Supported AI Clients
-
-GitLab MCP officially documents setup for Claude Code, VS Code, GitHub Copilot, Codex, Cursor, and JSON-based MCP clients.
-
-- OpenCode
-
-## When to Use
-
-Use GitLab MCP when an AI coding agent needs to interact with a GitLab instance.
-
-Good fits:
-
-- Reviewing, approving, and merging merge requests.
-- Browsing repositories, files, and commit history.
-- Creating and managing issues and to-do items.
-- Inspecting CI/CD pipelines and job logs.
-- Managing wiki pages, milestones, labels, and releases.
-- Automating GitLab workflows from an AI agent.
-
-Avoid using it outside of GitLab-related workflows.
-
-## Requirements
-
-- Node.js >= 18 and npm (for `npx` or global install), or Homebrew (for `brew install`).
-- A GitLab account with API access.
-- A Personal Access Token with appropriate scopes, or OAuth2 credentials.
-
-## Related Skills
-
-Relevant skills in this repository:
-
-- [`backend-best-practices`](../../skills/backend-best-practices/SKILL.md): backend best practices for API design, testing, debugging, performance, and security.
-- [`code-review`](../../skills/code-review/SKILL.md): reviewing merge request diffs.