Does it support result caching?

Absolutely. It demonstrates how to use NodeCache to store agent outputs based on input hashes, significantly reducing redundant API calls.

How does this skill help reduce Lindy AI latency?

It provides scripts to measure baseline metrics and offers specific patterns for concise instruction refactoring, which reduces token processing time.

How do I manage high-concurrency workloads?

The skill provides patterns using the p-limit library to throttle concurrent requests, preventing API throttling and optimizing overall throughput.

Can I implement streaming with this skill?

Yes, the skill includes implementation guides for transitioning from non-streaming to real-time streaming using the Lindy SDK runStream method.

Lindy AI Performance Tuning

Name: Lindy AI Performance Tuning
Author: jeremylongshore

byjeremylongshore

•

1,242

•

分析与监控

Optimizes Lindy AI agent latency and throughput through instruction refinement, caching, and streaming implementation.

Lindy Performance Tuning is a specialized skill designed to enhance the efficiency of Lindy AI agents in production environments. It provides a structured approach to identifying bottlenecks by measuring P95/P99 latency and throughput, followed by actionable optimization patterns. Developers can leverage this skill to refactor verbose instructions for faster processing, implement response streaming for better user experiences, and integrate robust caching and concurrency controls. It is an essential tool for scaling AI agent workflows where response time and resource management are critical.

主要功能

01Controlled concurrency management using p-limit

02Real-time response streaming implementation

031,242 GitHub stars

04NodeCache-based result caching patterns

05Automated latency and throughput benchmarking

06Prompt compression and instruction optimization

使用场景

01Scaling batch processing tasks across multiple Lindy agents

02Reducing response times in user-facing AI chat interfaces

03Resolving timeout issues in complex AI agent workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jeremylongshore/claude-code-plugins-plus-skills lindy-performance-tuning

For use in Claude.ai and ChatGPT

主要功能

01Controlled concurrency management using p-limit

02Real-time response streaming implementation

031,242 GitHub stars

04NodeCache-based result caching patterns

05Automated latency and throughput benchmarking

06Prompt compression and instruction optimization

使用场景

01Scaling batch processing tasks across multiple Lindy agents

02Reducing response times in user-facing AI chat interfaces

03Resolving timeout issues in complex AI agent workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jeremylongshore/claude-code-plugins-plus-skills lindy-performance-tuning

For use in Claude.ai and ChatGPT