agentsclimarketplace

Query product events

Skill altertable-ai/skills/skills/query-product-events

Queries product events and identities with SQL (event counts, properties, user activity, traits). Use to answer a question about tracked behavior or confirm new tracking arrives.From its SKILL.md

Install
npx -y skills add altertable-ai/skills --skill query-product-events

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 7 stars7 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.4 KB, as published. Nobody here has run it

Query Product Events

Use this skill with the Altertable MCP server.

Quick Start

  1. Call initialize before any other Altertable MCP tool.
  2. Use list_catalogs, then inspect the product_analytics catalog with get_catalog. Do not assume schemas, views, or columns.
  3. Choose the identity-resolved event view for user-level analysis and the raw ingestion table only when inspecting payload delivery.
  4. Validate SQL with validate_sql, then execute it with query_lakehouse.
  5. State the time range, environment, identity semantics, and any data-quality limitations with the result.

Read the canonical Product Analytics overview and query guide for the current data model and supported analysis paths.

When to Use This Skill

  • Counting or listing tracked product events
  • Inspecting event properties or recent user activity
  • Joining events to resolved identities or traits
  • Investigating whether new instrumentation is arriving
  • Analyzing product behavior with SQL
  • Comparing event activity across time, environment, plan, or another dimension

Use instrument-product-analytics when the task changes application code or sends new events. Prefer:

  • analyze-funnels for ordered conversion and drop-off
  • analyze-web-traffic for pageviews, sessions, referrers, UTM, device, or country
  • build-segments for defining or comparing cohorts
  • query-lakehouse for generic or cross-catalog SQL

Query Workflow

1. Establish Context

Call initialize first. Confirm the organization and environment before inspecting data.

Clarify:

  • the event or behavior being measured
  • the requested time range and timezone
  • whether the metric counts events, sessions, or distinct people
  • which environment is in scope
  • whether anonymous and aliased identities should be resolved

2. Discover the Current Schema

Use list_catalogs and get_catalog to discover the actual product_analytics schemas, tables, views, columns, and semantic definitions.

Do not copy a table path from an old query without checking it. Product Analytics commonly exposes both raw ingestion data and identity-resolved analytical views; choose based on the question rather than convenience.

When a field is stored in event properties, inspect its observed shape before casting or aggregating it.

3. Build a Bounded Query

  • Fully qualify tables as catalog.schema.table.
  • Add an explicit time filter.
  • Use LIMIT while inspecting rows or property shapes.
  • Use deterministic ORDER BY when results may be paginated.
  • Define whether counts are raw events or distinct resolved users.
  • Guard casts and ratios against malformed values, nulls, and division by zero.

Validate with validate_sql. Use explain_sql before a complex or expensive query, then run query_lakehouse.

4. Check the Result

Before drawing a conclusion:

  • compare the result with total volume or a nearby time range
  • check null and unknown identity rates when analyzing users
  • look for duplicated retries or abrupt ingestion gaps
  • distinguish missing events from genuine zero activity
  • mention incomplete recent data if ingestion may still be processing

When validating new instrumentation, confirm the event name, environment, expected properties, and identity context. Do not expose a user's sensitive properties in the response.

5. Present the Answer

Report:

  • the metric and unit
  • exact time range and timezone
  • event-count versus distinct-user semantics
  • filters and environment
  • important data-quality caveats

Offer SQL or a rendered insight when useful, but do not save an insight unless the user asks.

Common Pitfalls

  1. Querying before initialize
  2. Assuming a table or column name instead of inspecting the catalog
  3. Mixing raw and identity-resolved events in one metric
  4. Treating event count as user count
  5. Omitting the time range or environment
  6. Casting arbitrary property values without checking their shape
  7. Interpreting an ingestion outage as a behavioral drop
  8. Returning sensitive traits or event properties unnecessarily

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most product growth skills give

Counted across 694 of the 879 authors here whose files we hold, read 2026-09-06

  • Check for product marketing context firstin 49 of 694, across 20 files
  • Validate the why before building featuresin 18 of 694, across 4 files
  • Respond to every comment in real-timein 17 of 694, across 6 files
  • Structure launch marketing across three channel typesin 16 of 694, across 4 files
  • Recruit early users one-on-onein 13 of 694, across 2 files
  • Ask one question at a timein 13 of 694
  • Rank features using ICE scoringin 12 of 694, across 3 files
  • Identify primary conversion goalin 11 of 694, across 3 files
  • Identify traffic contextin 11 of 694, across 3 files
  • Evaluate headline effectivenessin 11 of 694, across 3 files
  • Check visual hierarchy and scannabilityin 11 of 694, across 3 files
  • Run product diagnosticsin 11 of 694, across 3 files

Said here and by no other author read

  • Call initialize before any other tool
  • Inspect the product analytics catalog
  • Choose identity resolved views for user analysis
  • Validate SQL before executing it
  • State the time range and environment
  • Fully qualify tables as catalog schema table

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.

Keep looking

Skills are one crate of 610,677. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.