skills.rest
boundless-xyz
boundless-xyzOFFICIAL
Explore all skills by this author →
GitHub

ops-telemetry-query

Query Boundless broker telemetry from Redshift and DuckDB Parquet archives.

195|172|Updated Sep 25, 2024
Search Tags:#analytics#telemetry#parquet#duckdb#redshift#boundless
One-click install
npx skills add https://github.com/boundless-xyz/boundless --skill ops-telemetry-query
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ops-telemetry-query
Source: https://github.com/boundless-xyz/boundless/tree/main/.claude/skills/ops-telemetry-query
Command: npx skills add https://github.com/boundless-xyz/boundless --skill ops-telemetry-query

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This internal capability enables Boundless operators and developers to quickly access and analyze broker telemetry data across live Redshift deployments and archived Parquet data, simplifying health and performance investigations.

Core Features & Use Cases

  • Route queries by time window to the live Redshift path (>2026-04-25) or the archive path (through 2026-04-24) using the DuckDB + Parquet workflow.
  • Query three telemetry tables: broker_heartbeats, request_evaluations, and request_completions for health, evaluations, and completions analyses.
  • Provide ready-made templates and guidance for environment credential discovery (REDSHIFT_URL or network_secrets.toml) and back-end selection to diagnose issues in prod and staging.

Quick Start

Identify the time window you need, export REDSHIFT_URL if querying live, and run the sample queries against telemetry views or the DuckDB archive path accordingly.

Frequently Asked Questions about ops-telemetry-query

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

FAQPage Schema
How do I query broker telemetry data stored in Parquet archives using DuckDB?▼

To query broker telemetry archives, use DuckDB to read Parquet files containing historical data through 2026-04-24, applying provided SQL templates against the telemetry schema to analyze request evaluations and completions.

What is the best way to diagnose broker health across prod and staging environments?▼

Diagnosing broker health requires querying the broker_heartbeats telemetry table using structured SQL templates, routing to live Redshift for recent data or DuckDB archives for historical records based on your time window.

Do I need to export REDSHIFT_URL to query live Boundless telemetry?▼

Yes, exporting REDSHIFT_URL is required for credential discovery to query live telemetry data in Redshift for records after 2026-04-25, or you can use network_secrets.toml for environment configuration.

Can I use DuckDB to analyze historical telemetry skip rates and proving times?▼

Yes, DuckDB queries Parquet archives to analyze historical telemetry skip rates and proving times from request_evaluations and request_completions tables for data through 2026-04-24.

How does time-based routing work when querying Boundless telemetry across different backends?▼

Time-based routing directs queries to live Redshift for data after 2026-04-25 and to DuckDB reading Parquet archives for data through 2026-04-24, ensuring distinct backends handle live versus historical telemetry.

What telemetry views are available under the telemetry schema for request completions analysis?▼

The telemetry schema provides views for broker_heartbeats, request_evaluations, and request_completions tables, offering structured query patterns to analyze request completions, evaluations, and broker health metrics.

Related Skills

Explore More Data & Analytics Skills→
Current-cmdCurrent-cmd

ops-telemetry-query

Query Boundless telemetry views for broker heartbeats, evaluations, and completions.

Community
Intermediate
devs6186devs6186

health

Compute a weighted 0-10 code health score from type, lint, test, and dead code checks.

Community
Advanced
Alex1980AlexAlex1980Alex

duckdb-analytics

Analyze JSONL and Parquet logs directly with DuckDB.

Community
Advanced
MuhammadUAMuhammadUA

sql-queries

Generate dialect-specific SQL for major warehouse platforms with built-in patterns and error handling guidance.

Community
Intermediate