Sentry Monitoring

Mode: Cognitive/Prompt-Driven — No standalone utility script; use via agent context.

Sentry Monitoring Skill

Overview

Provides 90%+ context savings vs raw Sentry API calls. Progressive disclosure by feature category: error tracking, performance monitoring, release management, and project configuration.

Requirements

• Sentry account with project configured
• SENTRY_AUTH_TOKEN environment variable (optional, for authenticated API calls)
• SENTRY_ORG environment variable (optional, defaults to first organization)
• SENTRY_PROJECT environment variable (optional, defaults to first project)

Tools (Progressive Disclosure)

Error Tracking

Performance Monitoring

Release Management

Project Configuration

Quick Reference

# List recent issues
curl -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/projects/$SENTRY_ORG/$SENTRY_PROJECT/issues/?query=is:unresolved"

# Get issue details
curl -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/issues/{issue_id}/"

# Resolve issue
curl -X PUT -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/issues/{issue_id}/" \
  -d '{"status": "resolved"}'

# List transactions
curl -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/organizations/$SENTRY_ORG/events/?field=transaction"

# Create release
curl -X POST -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/organizations/$SENTRY_ORG/releases/" \
  -d '{"version": "1.0.0", "projects": ["project-slug"]}'

# List projects
curl -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  "https://sentry.io/api/0/organizations/$SENTRY_ORG/projects/"

Configuration

Environment Variables

Getting Auth Token

1. Navigate to Sentry Settings → Account → API → Auth Tokens
2. Create new token with scopes: project:read, project:write, event:read
3. Set as environment variable: export SENTRY_AUTH_TOKEN=your_token_here

Security

⚠️ Never expose auth tokens in output ⚠️ Destructive operations (resolve-issue, ignore-issue, create-release, set-commits) require confirmation ⚠️ Use environment variables for credentials, never hardcode

Agent Integration

• devops (primary): Production monitoring, incident response
• incident-responder (primary): Error triage, issue resolution
• developer (secondary): Debugging, performance optimization
• qa (secondary): Test environment monitoring

Error Handling

If tool execution fails:

1. Verify SENTRY_AUTH_TOKEN is set: echo $SENTRY_AUTH_TOKEN
2. Check token permissions include required scopes
3. Verify organization and project slugs are correct
4. Review Sentry API rate limits (default: 3000 requests/minute)

Common Workflows

Incident Response

1. list-issues - Get recent unresolved errors
2. issue-details - Investigate specific issue
3. resolve-issue - Mark as resolved after fix deployed

Performance Optimization

1. list-transactions - Identify slow endpoints
2. transaction-summary - Analyze performance patterns
3. slow-queries - Find database bottlenecks

Release Management

1. create-release - Create new release version
2. set-commits - Associate commits with release
3. list-releases - Track release health

Troubleshooting

Related

• Sentry API Documentation: https://docs.sentry.io/api/
• Sentry Error Tracking: https://docs.sentry.io/product/issues/
• Sentry Performance: https://docs.sentry.io/product/performance/
• Sentry Releases: https://docs.sentry.io/product/releases/

Memory Protocol (MANDATORY)

Before starting: Read .claude/context/memory/learnings.md

After completing:

• New pattern -> .claude/context/memory/learnings.md
• Issue found -> .claude/context/memory/issues.md
• Decision made -> .claude/context/memory/decisions.md

ASSUME INTERRUPTION: If it's not in memory, it didn't happen.

OpenTelemetry + Grafana Stack

Auto-instrument with OTel, export to Sentry + Grafana simultaneously:

import { NodeSDK } from '@opentelemetry/sdk-node';
import { SentrySpanProcessor } from '@sentry/opentelemetry';
const sdk = new NodeSDK({ spanProcessor: new SentrySpanProcessor() });
sdk.start();

Continuous Profiling (Pyroscope)

# pyroscope.configure({ applicationName: 'api', serverAddress: 'http://pyroscope:4040' })

Grafana + Prometheus

services:
  prometheus: { image: prom/prometheus }
  grafana: { image: grafana/grafana }
  otel-collector: { image: otel/opentelemetry-collector-contrib }