Skill UI

This guide details the recommended pattern for managing per-project plugin settings and state within a development environment. Plugins store their configuration in a dedicated `.claude/plugin-name.local.md` file. This file utilizes YAML frontmatter for structured settings (like enabling flags, retry counts, and parameters) and a Markdown body for additional context or instructions. It provides techniques for reading these settings using Bash hooks, commands, and agents.

Guides on handling PostHog rate limits with exponential backoff, jitter, idempotent keys, queuing, and header monitoring to keep API calls reliable during throttling or retry scenarios.

Provides Replit-specific rate limiting guidance, adding exponential backoff with jitter, deterministic idempotency keys, queue throttling, and header monitoring so apps stay within quotas while retrying safely.

Guidance for applying production-ready Replit SDK patterns in TypeScript and Python, covering singleton clients, structured error handling, retry/backoff and validation so integrations stay resilient and teams share a standard.

Provides Retell AI SDK coding patterns for TypeScript and Python, including singleton clients, error wrapping, retry/backoff, validation, and logging so teams can implement reliable, idiomatic integrations quickly.

Guides API integration teams through retry logic helpers, outlining patterns, best practices, production-ready code, and validation for webhooks, SDKs, OAuth, and third-party APIs.

Guides building resilient Sentry integrations by outlining failure-aware init, retry/backoff, fallback logging, offline queues, circuit breakers, dual writers, and health checks so error tracking stays operational despite network or SDK issues.

Provides Speak language API rate-limiting helpers, including per-endpoint throttling, batch queues, and retry-on-429 handling to keep pronunciation assessments and conversation calls inside the documented quotas.

Guides Temporal workflow orchestration in the Python SDK, covering durable workflow patterns, saga/distributed transactions, async/sync activity models, retry and timeout policies, signal/query handling, type-safe APIs, and testing/deployment best practices.

Automates guidance for configuring test retries in automation suites, walking through setup, best practices, and frameworks such as Jest or pytest so teams can stabilize flaky runs.

Provides Vast.ai-specific rate limit handling with backoff plus jitter, Idempotency key support, and queue-based throttling so API clients can retry safely under 429 responses and monitor headers for throughput control.

Provides guidance for handling Vercel rate limits, backoff, and idempotency so APIs can retry reliably, respect limit headers, and avoid duplicate requests.