skills.rest
Lightricks
LightricksOFFICIAL
Explore all skills by this author →
Website𝕏@lightricksGitHub

be-cost-monitoring

Analyze backend GPU costs for LTX API and LTX Studio.

Updated Aug 27, 2026
Search Tags:#monitoring#infrastructure#gpu#bigquery#anomaly-detection#cost
One-click install
npx skills add https://github.com/Lightricks/ltx-analytics-agents --skill be-cost-monitoring
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: be-cost-monitoring
Source: https://github.com/Lightricks/ltx-analytics-agents/tree/main/agents/monitoring/be-cost
Command: npx skills add https://github.com/Lightricks/ltx-analytics-agents --skill be-cost-monitoring

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides structured guidance and tooling to monitor backend GPU costs for LTX API and LTX Studio, enabling teams to spot trends, spikes, and cost drivers before they impact budgets.

Core Features & Use Cases

  • Break down costs by API endpoint, model, and org for API usage.
  • Break down costs by Studio process/workspace and track utilization versus idle costs.
  • Detect anomalies, compare periods, and alert on budget breaches.

Quick Start

Provide a baseline cost analysis for the last 7 days and surface top cost drivers to prioritize investigation.

Frequently Asked Questions about be-cost-monitoring

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

FAQPage Schema
How do I monitor GPU costs for API and Studio usage?▼

GPU cost monitoring analyzes backend costs for LTX API and LTX Studio to identify trends, spikes, and cost drivers. It breaks down API costs by endpoint, model, and org, and Studio costs by process and workspace.

What is the best way to detect cost anomalies in BigQuery GPU infrastructure data?▼

GPU cost anomaly detection compares daily, weekly, and monthly periods to spot unexpected spikes. It applies cost category 'inference' for per-request usage and uses SAFE_DIVIDE on BigQuery partitioned data to accurately surface cost drivers.

Can I break down GPU spend by workspace and process for LTX Studio?▼

Yes, GPU spend breakdown for LTX Studio tracks costs by process and workspace. It compares utilization against idle costs to help teams identify inefficient resource allocation and optimize overall spend.

How do I generate a baseline GPU cost analysis for the last 7 days?▼

Generate a 7-day baseline cost analysis by querying BigQuery dt partition data for API and Studio. This surfaces top cost drivers, applies the 'inference' category for per-request costs, and prioritizes investigation into spending spikes.

Does this Skill require reading metric standards before querying cost data?▼

Yes, analyzing GPU costs requires reading shared/metric-standards.md and schema files first. The process mandates filtering on the dt partition, using SAFE_DIVIDE for calculations, and categorizing per-request costs as 'inference'.

Related Skills

Explore More Data & Analytics Skills→
abdullah1854abdullah1854
1

sub-agent-invocation

Route tasks to specialized sub-agents with context isolation for parallel work.

Community
Advanced
aalmadaaalmada
15

scaffold-composite-projection

Build a Composite Projection in Marten to chain multiple projections.

Community
Advanced
nimadorostkarnimadorostkar
25

aws-cost-optimization

Analyze AWS Cost Explorer data to identify and mitigate cost drivers.

Community
Advanced
lucidashlucidash
5

good-morning

Aggregates GitHub, Notion, and Slack activity into a prioritized daily task list and Slack-ready payload.

Community
Intermediate