session-trends

Summarize windowed session metrics trends from a read-only ledger file.

7|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/slbug/claude-ruby-grape-rails --skill session-trends
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-trends
Source: https://github.com/slbug/claude-ruby-grape-rails/tree/main/.claude/skills/session-trends
Command: npx skills add https://github.com/slbug/claude-ruby-grape-rails --skill session-trends

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Summarize distributed session metrics into clear, windowed trend reports so contributors can monitor adoption, friction, and opportunity without manual aggregation and ad-hoc scripting.

Core Features & Use Cases

  • Windowed Aggregation: Compute 7d, 30d, and all-window aggregates from a session metrics ledger for quick comparison.
  • Scoped Filtering: Support project- and provider-scoped analysis to avoid misleading mixed-provider summaries.
  • Comparison and Guidance: Optionally compare computed aggregates against a provided notes file and suggest cautious next actions for monitoring and investigation.
  • Use Case: Prioritize investigation when rising friction or missed opportunities appear in recent windows while ensuring provider mixing is called out.

Quick Start

Run the session-trends skill to summarize the .claude/session-metrics/metrics.jsonl ledger for the last 30 days and present a readable trend report.

Frequently Asked Questions about session-trends

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

FAQPage Schema
How do I summarize session metrics into windowed trend reports?

You can compare session metric aggregates by using the --compare flag against a provided notes file. This helps identify rising friction or missed opportunities in recent windows while suggesting cautious next actions for investigation.

Can I filter session trend analysis by provider and project scope?

Yes, you can filter session trend analysis using --provider and --project flags. Scoped filtering prevents misleading mixed-provider summaries and enables focused provider comparisons or project-specific adoption monitoring.

What is the best way to monitor session adoption and friction without ad-hoc scripting?

The best way to monitor adoption and friction without ad-hoc scripting is to use automated trend aggregation on a canonical metrics ledger. This computes time-series aggregates directly, replacing manual data extraction with consistent windowed summaries.

Do I need a specific metrics ledger format to compute session trend aggregates?

Yes, you need a read-only .claude/session-metrics/metrics.jsonl ledger file to compute session trend aggregates. The skill relies on this canonical JSONL format and the compute-metrics.py script for accurate aggregation.

Why does my session trend summary show misleading mixed-provider data?

Session trend summaries show misleading mixed-provider data when provider-scoped filters are not applied. Using --provider flags isolates specific provider metrics, ensuring accurate provider comparisons and preventing skewed aggregate calculations.