Claude Code Plugins

Community-maintained marketplace

Feedback

integration-patterns

@romiluz13/cc10x
8
0

Focused integration design and reliability guidance. Thin wrapper around the Integration section of design-patterns with a compact checklist and examples. Use when designing integrations, planning service contracts, implementing retry strategies, handling integration failures, or ensuring reliable cross-service communication. Used by integration-verifier, planning-workflow, and debug workflows when integrations fail.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name integration-patterns
description Focused integration design and reliability guidance. Thin wrapper around the Integration section of design-patterns with a compact checklist and examples. Use when designing integrations, planning service contracts, implementing retry strategies, handling integration failures, or ensuring reliable cross-service communication. Used by integration-verifier, planning-workflow, and debug workflows when integrations fail.

Integration Patterns (Focused)

Progressive Loading Stages

Stage 1: Metadata

  • Purpose: Make integrations reliable, observable, and resilient
  • When: Designing or verifying service/service or service/third-party interactions
  • Core Rule: Integrations fail; design for failure and recovery

Stage 2: Quick Reference

Integration Checklist

Reliability
- [ ] Timeouts set (client + server)
- [ ] Retries with exponential backoff and jitter
- [ ] Idempotency keys for retryable operations
- [ ] Circuit breaker around downstream dependencies

Resilience & Consistency
- [ ] Dead-letter queue or compensating actions
- [ ] Outbox/inbox where needed (avoid dual writes)
- [ ] Event/version compatibility; schema evolution plan

Observability & Ops
- [ ] Correlated logs/trace IDs
- [ ] Dashboards and alerts for SLOs
- [ ] Runbooks for common failures

Examples

Exponential Backoff (TS pseudo)

for (let i = 0; i < max; i++) {
  const delay = Math.min(base * 2 ** i + jitter(), cap);
  try { return await call(); } catch (e) { await sleep(delay); }
}
throw new Error('exhausted retries');

Circuit Breaker Settings

  • Failure threshold, half-open probe, cooldown window

Stage 3: Verification Procedure

  1. Simulate timeouts and verify retry/backoff/circuit breaker behavior
  2. Verify idempotency on repeated calls
  3. Check logs/traces contain correlation IDs and diagnoses
  4. Validate DLQ/outbox handling and runbooks

Stage 4: Outputs

  • Integration Verification Checklist with results and logs
  • Incidents and runbook gaps captured

Stage 5: Links

  • See also: design-patterns (Integration section), log-analysis-patterns, systematic-debugging, risk-analysis