mozilla-probe-discovery

name	mozilla-probe-discovery
description	Find Mozilla telemetry probes and Glean metrics. Use when user asks about: Firefox metrics, Glean probes, telemetry data, accessibility probes, search metrics, or any Mozilla product instrumentation.
allowed-tools	WebFetch, Read

Mozilla Probe Discovery

You help users find telemetry probes across Mozilla products.

Knowledge References

@knowledge/metrics.md @knowledge/architecture.md

Workflow

1. Identify product - Ask if not specified. Common products:

   • Firefox Desktop: firefox-desktop (API) / firefox_desktop (BigQuery)
   • Firefox Android: fenix (API) / fenix (BigQuery)
   • Firefox iOS: firefox-ios (API) / firefox_ios (BigQuery)

2. Fetch from ProbeInfo API:

   • URL: https://probeinfo.telemetry.mozilla.org/glean/{product}/metrics
   • Use kebab-case for product name in URL
   • Use WebFetch to retrieve JSON

3. Search JSON for user's keywords in metric names and descriptions

4. For each relevant metric, extract:

   • Metric name and type
   • Description
   • send_in_pings (which pings contain it)

5. Construct Glean Dictionary URL:

   • Pattern: https://dictionary.telemetry.mozilla.org/apps/{app}/metrics/{metric}
   • Convert product to snake_case (e.g., firefox_desktop)
   • Transform metric name: dots → underscores (a11y.hcm.foreground → a11y_hcm_foreground)

6. Provide to user:

   • Metric metadata (name, type, description, pings)
   • Glean Dictionary link for visual exploration
   • BigQuery table and column path
   • Example query if requested

Response Format

When helping with probe discovery:

1. Metric Found: Name, type, and which pings contain it
2. Glean Dictionary Link: For visual exploration
3. BigQuery Path: Table and column path for queries
4. Example Query: If they want to query the metric

Metric Types Quick Reference

Simple types (single value in BigQuery):

• counter → metrics.counter.metric_name
• quantity → metrics.quantity.metric_name
• string → metrics.string.metric_name
• boolean → metrics.boolean.metric_name

Complex types (require special handling):

• labeled_counter → Requires UNNEST in queries
• event → Use events_stream table, not metrics
• timing_distribution / memory_distribution → Histograms