daily-brief

Generates a daily briefing of metric changes, experiments, and feedback from Amplitude analytics data.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/amplitude/amplitude-copilot-plugin --skill daily-brief-amplitude
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: daily-brief
Source: https://github.com/amplitude/amplitude-copilot-plugin/tree/main/skills/plugins/amplitude/daily-brief
Command: npx skills add https://github.com/amplitude/amplitude-copilot-plugin --skill daily-brief-amplitude

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Product teams waste time manually checking dashboards every morning to figure out what changed overnight. This Skill automatically scans your Amplitude instance and delivers a concise, prioritized briefing of anomalies, experiment results, and user feedback from the last 1-2 days. ## Core Features & Use Cases - Automated anomaly detection: Queries charts across your org's most-viewed dashboards, ranks day-over-day changes, and filters out false positives like partial-day artifacts and day-of-week effects. - Root cause analysis: Cross-references deployments, experiments, and feedback to explain why metrics moved, with segment breakdowns for the biggest findings. - Persona-calibrated briefings: Adapts language and priorities for executives, PMs, analysts, growth, and engineering roles. - Use Case: A PM asks "anything I should know today?" and receives a 400-600 word memo highlighting that a checkout experiment hit significance overnight and search adoption dropped 15%, each with a concrete recommended action. ## Quick Start Ask your AI assistant: "Give me my daily download from Amplitude."

Frequently Asked Questions about daily-brief

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

FAQPage Schema
How do I get a daily summary of my Amplitude metrics?

Ask your AI assistant for a daily brief or morning briefing. The skill scans your org's top dashboards and charts, compares today and yesterday against the trailing 5 days, and delivers a narrative summary of the biggest changes with recommended actions.

How does the daily brief detect anomalies in product metrics?

It queries charts at daily granularity over 7 days, computes day-over-day deltas, and flags metrics deviating more than 15% from the recent average. It then filters false positives like partial-day data, day-of-week effects, and incomplete retention cohorts.

Can the daily brief explain why a metric changed?

Yes. It cross-references recent deployments, experiment status changes, and user feedback to hypothesize causes. For the top 2-3 findings, it runs segment breakdowns by platform, country, or plan tier to isolate where the change concentrates.

Does the daily brief work if I have no dashboards in Amplitude?

It falls back to broad searches for any available charts or events and builds context from whatever exists. It will notify you that setup is limited and suggest creating a key metrics dashboard.

Why does the daily brief warn about partial-day data?

Today's data is incomplete, so comparing raw totals to yesterday's full day produces misleading drops. The skill compares per-hour pacing against yesterday's same-hour rate and explicitly notes when figures are partial.