langchain-rate-limits

Quick Review

Excellent skill providing comprehensive rate limiting and retry strategies for LangChain applications. The description clearly communicates when to invoke the skill with specific trigger phrases. Task knowledge is strong with concrete code examples covering built-in retries, tenacity integration, custom rate limiters (sync/async), batch processing, and fallback patterns. Structure is logical with clear steps from basic to advanced patterns, helpful error handling table, and provider-specific rate limit reference. Novelty is moderate-to-good: while retry logic is common, the LangChain-specific implementations, async patterns, and token bucket rate limiter provide meaningful value over generic solutions. The skill would help an agent implement production-ready rate limiting without extensive trial-and-error, reducing token costs for complex quota management scenarios. Minor improvement areas: could benefit from more discussion of token-per-minute (TPM) limiting vs request-per-minute (RPM), and the rate limiter could track both metrics simultaneously.