observability

Diagnose Navigator telemetry and durable-execution failures using identifier-only spans, metrics, and logs.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/neon-law-foundation/navigator --skill observability-neon-law-foundation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: observability
Source: https://github.com/neon-law-foundation/navigator/tree/main/.claude/skills/observability
Command: npx skills add https://github.com/neon-law-foundation/navigator --skill observability-neon-law-foundation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps engineers observe Navigator safely and diagnose durable-execution failures without leaking client content into telemetry.

Core Features & Use Cases

  • Unified Telemetry Setup: Standardizes how Navigator binaries initialize logs, traces, and metrics through one shared telemetry entry point.
  • Safe Instrumentation Rules: Guides you to record identifiers and counts instead of sensitive content when adding spans, metrics, or log fields.
  • Operational Debugging: Supports troubleshooting missing periodic jobs, silent triggers, unready workloads, and other workflow failures across logs, BigQuery, and the Restate console.
  • Use Case: A trigger job stops firing overnight, and this Skill helps you trace the failure from navigator doctor to telemetry signals and the underlying service or scheduling issue.

Quick Start

Use the observability skill to investigate a missing trigger by checking navigator doctor, then correlating logs, traces, and Restate invocations.

Frequently Asked Questions about observability

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

FAQPage Schema
How do I debug missing periodic jobs and silent triggers in durable execution?

Debug missing periodic jobs and silent triggers by running navigator doctor, then correlating logs, traces, and Restate invocations to trace the failure from telemetry signals to the underlying service or scheduling issue.

What is the safest way to instrument telemetry for service binaries without leaking content?

Safe telemetry instrumentation requires recording only identifiers and counts instead of sensitive content when adding spans, metrics, or log fields to service binaries.

How do I set up unified telemetry initialization for logs, traces, and metrics?

Set up unified telemetry initialization through one shared telemetry entry point to standardize how binaries initialize logs, traces, and metrics before exporting.

Does telemetry export work with BigQuery and the OTel Collector?

Yes, telemetry export works with BigQuery and the OTel Collector when OTLP export is configured, allowing you to deliver logs, traces, and metrics to either destination for operational debugging.

Why is my workflow handler instrumentation not showing up in traces?

Workflow handler instrumentation may not show up in traces if it lacks identifier-only spans, uses sensitive content instead of identifiers, or lacks shared telemetry initialization with OTLP export configured.

What's the best way to diagnose unready workloads and workflow failures across logs and the Restate console?

Diagnose unready workloads and workflow failures by correlating telemetry signals across logs, BigQuery, and the Restate console, using navigator doctor investigations to pinpoint the root cause.