speed-matcher

Optimize model selection logic in the Select harness to prioritize lower-latency candidates.

Updated May 11, 2026
One-click install
npx skills add https://github.com/AesopScott/mojo --skill speed-matcher
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: speed-matcher
Source: https://github.com/AesopScott/mojo/tree/main/harnesses/skills/speed-matcher
Command: npx skills add https://github.com/AesopScott/mojo --skill speed-matcher

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses latency and throughput bottlenecks in the Select harness by providing a structured framework to weight model selection toward faster candidates without sacrificing task quality.

Core Features & Use Cases

  • Performance Tuning: Adjusts the weighting logic for model selection based on latency, capability, and cost.
  • Harness Control: Provides a systematic runbook for inspecting and modifying the Select harness configuration.
  • Use Case: When a specific API route is experiencing high latency or queue stalls, use this Skill to implement a performance-focused rule that prioritizes faster models for non-critical tasks.

Quick Start

Use the speed-matcher skill to analyze the current Select harness configuration and propose a performance-oriented rule change to reduce latency for the identified bottleneck.

Frequently Asked Questions about speed-matcher

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I optimize model selection for lower API latency?

To optimize model selection for lower latency, you adjust the weighting logic in the Select harness configuration to prioritize faster candidates. This balances execution speed against task quality to resolve throughput bottlenecks.

What is harness weighting for performance tuning?

Harness weighting for performance tuning is a mechanism that adjusts model selection logic based on latency, capability, and cost. It provides a structured framework to prioritize faster candidates without sacrificing output quality.

Can I prioritize faster AI models for non-critical tasks only?

Yes, you can prioritize faster models for non-critical tasks by implementing a performance-focused rule change. This targets specific API routes experiencing high latency or queue stalls while preserving quality where needed.

Do I need performance logs to reduce AI routing latency?

Yes, reducing AI routing latency requires access to harness configuration files, performance logs, and defined success metrics. These inputs are necessary to validate the optimization and ensure throughput improvements.

What's the best way to resolve queue stalls in a model selection harness?

The best way to resolve queue stalls in a model selection harness is to apply a structured runbook to inspect and modify configuration rules. This shifts routing logic to prioritize lower-latency candidates for affected tasks.

Why does prioritizing faster models risk task quality?

Prioritizing faster models risks task quality because speed-focused candidates may lack the full capability profile of slower alternatives. The Skill mitigates this by weighting selection based on latency, capability, and cost together.