Tool Integrations

StackPilot integrates deeply with your existing development and monitoring toolchain to provide comprehensive incident analysis and automated response.

Core Integrations

GitHub Integration

Purpose: Code analysis, commit correlation, and automated PR generation

Required Permissions:

• Read access to code and commit history
• Write access to create pull requests and comments
• Read access to deployment status and releases

Setup Process:

1. Navigate to Connections in your team workspace
2. Click Connect GitHub
3. Authorize the StackPilot GitHub App
4. Select repositories to monitor

What StackPilot Does:

• Analyzes recent commits when incidents occur
• Correlates stack traces with specific code changes
• Generates pull requests with proposed fixes
• Tracks deployment correlation with incident timing

PagerDuty Integration

Purpose: Automatic incident detection and Slack channel joining

Setup Process:

1. In PagerDuty, create a new webhook for your service
2. Set webhook URL to your StackPilot webhook endpoint
3. Configure incident triggers (Critical, High priority incidents)
4. Add StackPilot bot to your incident response Slack channels

Incident Workflow:

1. PagerDuty fires alert → StackPilot receives webhook
2. StackPilot joins associated Slack channel
3. Posts initial analysis within 30 seconds
4. Provides suggested investigation steps

Sentry Integration

Purpose: Error tracking, stack trace analysis, and exception correlation

Setup Process:

1. Generate a Sentry auth token with project read access
2. Add token to StackPilot Connections
3. Select projects to monitor

Capabilities:

• Real-time error detection and analysis
• Stack trace correlation with git commits
• Error frequency and user impact analysis
• Automatic error grouping and pattern recognition

Datadog Integration

Purpose: Log aggregation, metrics analysis, and automated log querying

Setup Process:

1. Create Datadog API and App keys
2. Grant log read and metrics read permissions
3. Configure in StackPilot Connections

Features:

• Automated log query generation based on incidents
• Metric correlation (CPU, memory, response time)
• Service dependency mapping
• Performance regression detection

Slack Integration

Purpose: Team communication, notifications, and interactive incident response

Setup Process:

1. Install StackPilot Slack App to your workspace
2. Invite StackPilot bot to relevant channels
3. Configure notification preferences

Capabilities:

• Automatic incident channel joining
• Interactive Q&A during investigations
• Timeline updates and status notifications
• Team member mentions and escalations

Advanced Integrations

Kubernetes Integration

Monitor deployment health and correlate incidents with cluster events:

• Pod restart detection
• Resource constraint analysis
• Deployment rollback suggestions
• Service mesh traffic analysis

AWS CloudWatch Integration

Comprehensive cloud infrastructure monitoring:

• Lambda function error analysis
• RDS performance correlation
• Load balancer health checks
• Auto Scaling event correlation

New Relic Integration

Alternative monitoring solution with deep application insights:

• Application performance monitoring
• Database query analysis
• Browser error tracking
• Infrastructure correlation

Integration Workflow Example

Sample Integration Response

When an incident occurs, StackPilot automatically:

1. Receives PagerDuty webhook with incident details
2. Queries Sentry for related error details and stack traces
3. Analyzes GitHub for recent commits in affected code paths
4. Searches Datadog logs for relevant error patterns
5. Posts to Slack with initial analysis:

🚨 **Incident Alert: API 500 Errors**

**Root Cause Analysis** (87% confidence):
- **Suspect Commit**: #7d1e9 by @amy 8 mins ago
- **Error Pattern**: NullPointerException in ImageProcessor.java:122
- **Impact**: 47 users affected, 156 errors in last 10 minutes

**Suggested Investigation**:
• `service:api status:error "ImageProcessor" @timestamp:[now-15m TO now]`
• Check recent deployment logs for v3.8.1
• Review PR #342 changes to optimize() function

**Next Steps**:
1. Validate null check in ImageProcessor.optimize()
2. Consider immediate rollback if impact increases
3. Test fix in staging environment

*Type "help" for more investigation options*

Security & Permissions

OAuth Scopes

StackPilot requests minimal necessary permissions:

• GitHub: repo (read), pull_requests (write), deployments (read)
• Slack: channels:history, chat:write, files:read
• Datadog: logs_read, metrics_read
• Sentry: project:read, org:read

Data Access Policies

• Log Data: Only accessed during active incidents
• Code Analysis: Limited to public repositories and connected private repos
• Retention: Incident data retained per team policy (30-365 days)
• Encryption: All data encrypted in transit and at rest

Webhook Security

• Signature Verification: All webhooks verified with shared secrets
• IP Allowlisting: Restrict webhook sources to trusted IPs
• Rate Limiting: Prevent webhook flooding and abuse
• Audit Logging: Complete trail of all integration activity

Troubleshooting Integrations

Common Issues

GitHub PR Creation Fails:

• Verify bot has write permissions to repository
• Check if branch protection rules block automated PRs
• Ensure repository is not archived or read-only

Slack Notifications Missing:

• Confirm bot is invited to incident channels
• Check notification preferences in team settings
• Verify Slack app permissions include chat:write

Datadog Queries Timeout:

• Review API rate limits and key permissions
• Optimize log retention and indexing
• Consider upgrading Datadog plan for higher limits

PagerDuty Webhooks Not Received:

• Validate webhook URL configuration
• Check firewall and network access rules
• Test webhook delivery in PagerDuty interface

By properly configuring these integrations, StackPilot becomes a powerful central hub for your incident response workflow, automatically correlating data across tools and providing intelligent insights to accelerate resolution.