review-agent-insights

Synthesizes recent Amplitude AI agent results into a ranked findings digest.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Amplitude's AI agents generate analyses across many agent types, making it hard to track what the AI has found, whether findings are still current, and which ones matter most. This Skill consolidates all recent agent results into one prioritized narrative. ## Core Features & Use Cases - Cross-Agent Aggregation: Queries every available agent type in get_agent_results in parallel and merges duplicate findings across agents. - Freshness Validation: Scores findings by age and cross-references deployments to flag issues that may already be resolved. - Ranked Digest Output: Groups findings by product theme, ranks by impact and evidence strength, and highlights coverage gaps. - Use Case: A product manager asks "what has the AI found this week?" and receives a forwardable digest of the top findings with recommended actions and follow-up prompts. ## Quick Start Ask your AI assistant to review all AI agent insights from Amplitude over the past week and summarize the most important findings.

Frequently Asked Questions about review-agent-insights

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

FAQPage Schema
How do I review all AI agent insights in Amplitude?

Ask your AI assistant to review recent agent insights. The Skill queries every agent type available in get_agent_results in parallel, filters by a time window (default 7 days), and synthesizes the results into a ranked digest grouped by product theme.

How do I check if Amplitude AI findings are still relevant?

The Skill validates freshness by checking each finding's creation date and cross-referencing get_deployments. Findings older than 14 days are excluded, and issues with a subsequent deployment are flagged as potentially resolved.

Can I filter Amplitude agent insights to a specific product area?

Yes. Mention a focus area like onboarding or checkout, and the Skill adds a matching query parameter to every agent type call, presenting only findings relevant to that area.

What happens if no Amplitude agent results are found?

The Skill reports that no recent analyses were found, lists every agent type that was queried, and suggests which agents to run on which dashboards or flows to generate new insights.

Does this Skill handle duplicate findings from multiple agents?

Yes. When the same problem surfaces from multiple agent types, the Skill merges them into a single finding with multi-agent evidence, which also raises its ranking in the final digest.