Transform your development workflow with MCP (Model Context Protocol) servers that seamlessly connect GitHub Copilot to your GitHub and Jira instances. Reduce context switching by 70% and triple your productivity.
- Context Overload: Automatically loads issue context without manual copying
- Repetitive Tasks: Automates PR creation, ticket updates, and documentation
- Mental Fatigue: Preserves cognitive bandwidth for actual problem-solving
- Multi-Tool Friction: Unifies GitHub and Jira workflows through natural language
- Docker installed and running
- GitHub Personal Access Token (Create one)
- Jira API Token (for Jira integration)
- VS Code 1.101+ with GitHub Copilot
- Clone the repository:
git clone https://github.com/yourusername/mcp-server-integration
cd mcp-server-integration- Create environment file:
cat > ~/.mcp/.env << EOF
GITHUB_PERSONAL_ACCESS_TOKEN=ghp_your_token_here
JIRA_PERSONAL_TOKEN=your_jira_token
JIRA_URL=https://your-domain.atlassian.net
EOF- Add to VS Code settings.json:
{
"chat.mcp.enabled": true,
"mcp": {
"servers": {
"github": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"--env-file", "${HOME}/.mcp/.env",
"-e", "GITHUB_PERSONAL_ACCESS_TOKEN",
"ghcr.io/github/github-mcp-server"
]
},
"jira": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"--env-file", "${HOME}/.mcp/.env",
"-e", "JIRA_URL",
"-e", "JIRA_PERSONAL_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
]
}
}
}
}- Enable Agent Mode in GitHub Copilot Chat and start using!
- 📁 Repository Management: Browse code, search files, analyze commits
- 🐛 Issue & PR Automation: Create, update, manage issues and PRs
- 🔄 CI/CD Intelligence: Monitor Actions, analyze failures
- 🔒 Security Scanning: Review security findings and Dependabot alerts
- 📝 Smart Issue Updates: Natural language ticket updates
- 🔍 Context Loading: Auto-fetch issue details into AI context
- 🏷️ Project Navigation: Search and filter issues across projects
- 🔗 Cross-Reference: Link PRs to tickets automatically
graph LR
A[VS Code/Copilot] -->|MCP Protocol| B[MCP Servers]
B --> C[GitHub MCP]
B --> D[Jira MCP]
C -->|API| E[GitHub]
D -->|API| F[Jira]
style A fill:#e1f5fe
style B fill:#fff3e0
style C fill:#f3e5f5
style D fill:#f3e5f5
| Component | Description | Technology |
|---|---|---|
| Transport Layer | Stdio/HTTP communication | JSON-RPC |
| Authentication | PAT/OAuth token management | Environment variables |
| Tool Registry | Modular tool organization | Go interfaces |
| API Clients | Service-specific wrappers | REST/GraphQL |
| Container Runtime | Isolated execution | Docker Alpine |
# GitHub Configuration
GITHUB_PERSONAL_ACCESS_TOKEN=ghp_xxxx
GITHUB_HOST=https://github.com # Or your enterprise URL
GITHUB_TOOLSETS=repos,issues,pull_requests,actions
GITHUB_READ_ONLY=false # Set true for safe mode
# Jira Configuration
JIRA_URL=https://your-domain.atlassian.net
JIRA_PERSONAL_TOKEN=xxxx
JIRA_SSL_VERIFY=trueEnable only what you need for better performance:
# Minimal setup
GITHUB_TOOLSETS="context,repos,issues"
# Full setup
GITHUB_TOOLSETS="all"
# Dynamic discovery (recommended)
GITHUB_DYNAMIC_TOOLSETS=1| Toolset | Description | Tools Included |
|---|---|---|
context |
User profile & teams | get_me, get_teams |
repos |
Repository operations | create_branch, get_commits, search_code |
issues |
Issue management | create_issue, update_issue, add_comment |
pull_requests |
PR workflows | create_pr, merge_pr, request_review |
actions |
CI/CD operations | run_workflow, get_logs, list_runs |
code_security |
Security scanning | get_alerts, list_vulnerabilities |
"Include context from JIRA-1234"
Copilot automatically fetches issue details and related information.
"Create a PR for the changes we discussed, linking to JIRA-1234"
Generates PR with description, links issue, and updates Jira.
"Find all critical bugs in the auth module and suggest fixes"
Searches issues, analyzes code, and proposes solutions.
"Why did the last deployment fail?"
Retrieves workflow logs, identifies errors, suggests fixes.
-
Token Management
- Never commit tokens to version control
- Use minimal required scopes
- Rotate tokens regularly
- Store in secure environment files
-
Access Control
- Use read-only mode for exploration
- Limit toolsets to required functions
- Review agent actions before confirmation
-
Network Security
- Use HTTPS for all connections
- Verify SSL certificates
- Restrict to allowed domains only
| Metric | Before MCP | After MCP | Improvement |
|---|---|---|---|
| Context Loading | 45% of time | 5% of time | 90% reduction |
| Issue Resolution | 2-3 hours | 30-45 mins | 75% faster |
| Daily Throughput | 2-3 issues | 8-10 issues | 3-4x increase |
| Documentation | 20% of time | Automated | 100% automated |
We welcome contributions! See CONTRIBUTING.md for guidelines.
# Build from source
go build -o mcp-server cmd/server/main.go
# Run tests
go test ./...
# Build Docker image
docker build -t mcp-server:local .- Define tool in
tools/directory - Add to appropriate toolset
- Update parameter schemas
- Add tests
- Submit PR
| Issue | Solution |
|---|---|
| "Connection refused" | Ensure Docker is running |
| "Authentication failed" | Check token permissions and expiry |
| "Tool not found" | Enable required toolset |
| "Rate limit exceeded" | Implement caching or reduce frequency |
# Enable verbose logging
MCP_VERBOSE=true docker run ...
# Test connection
docker run --rm ghcr.io/github/github-mcp-server --test- GitLab MCP server support
- Slack integration
- Linear.app support
- Custom tool builder UI
- Performance analytics dashboard
- Multi-workspace management
- Terraform/IaC integrations
- Anthropic for the MCP protocol
- GitHub for the official MCP server
- Community contributors and testers
MIT License - see LICENSE file
Built with ❤️ for the developer community
If this helped your workflow, please star ⭐ the repo and share your experience!