Agent Skills: Latency Tracker

Per-call and aggregated latency tracking for MEV infrastructure. Use when implementing performance monitoring or debugging slow operations. Triggers on: latency, timing, performance, slow, speed, instrumentation.

UncategorizedID: aiskillstore/marketplace/latency-tracker

Install this agent skill to your local

pnpm dlx add-skill https://github.com/aiskillstore/marketplace/tree/HEAD/skills/barissozen/latency-tracker

Skill Files

Browse the full folder contents for latency-tracker.

Download Skill

Loading file tree…

skills/barissozen/latency-tracker/SKILL.md

Skill Metadata

Name
latency-tracker
Description
"Per-call and aggregated latency tracking for MEV infrastructure. Use when implementing performance monitoring or debugging slow operations. Triggers on: latency, timing, performance, slow, speed, instrumentation."

Latency Tracker

Per-call and aggregated latency tracking for MEV infrastructure.

When to Use

  • Implementing performance monitoring
  • Debugging slow operations
  • Adding instrumentation to code paths
  • Tracking end-to-end latency
  • Setting up alerting thresholds

Workflow

Step 1: Define Span Hierarchy

Structure spans from e2e_flow down to individual calls.

Step 2: Instrument Code

Wrap operations in tracker.span() calls.

Step 3: Set Alert Thresholds

Configure alerts for latency exceeding expected ranges.


Span Hierarchy

e2e_flow (root) ├── rust_hotpath_call (5-15ms) │ └── rpc_eth_call (5-20ms) ├── tx_submit └── tx_confirm (1-15s)

Usage

const e2e = tracker.startE2E('liq');

await e2e.span('rust_call', async () => {
  return await callRust();
});

e2e.complete({ success: true });

Alert Thresholds

| Span | Expected | Alert | |------|----------|-------| | rust_hotpath | 5-15ms | >30ms | | rpc_eth_call | 5-20ms | >50ms | | e2e_to_submit | 10-50ms | >100ms |