insights

Computes session analytics, learning patterns, and correction trends from project memory and git history.

1|Updated May 6, 2026
One-click install
npx skills add https://github.com/surfingalien/FinSurfing --skill insights-surfingalien
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: insights
Source: https://github.com/surfingalien/FinSurfing/tree/main/.claude/skills/insights
Command: npx skills add https://github.com/surfingalien/FinSurfing --skill insights-surfingalien

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Developers using AI coding assistants accumulate learnings and corrections across sessions but have no visibility into whether those patterns are improving, stagnating, or being forgotten. This Skill turns raw session history and project memory into actionable analytics. ## Core Features & Use Cases - Session Summary: Reports duration, files modified, corrections applied, learnings captured, and context usage for the current session. - Learning Analytics: Breaks down learnings by category, ranks most-applied patterns, and flags stale learnings never used. - Correction Heatmap: Visualizes correction counts by category and highlights recurring corrections that should become permanent rules. - Productivity Metrics: Tracks average session length, edits per session, correction rate trends, and learning capture rate over recent sessions. - Use Case: After a week of coding sessions, ask for a stats dashboard to discover that testing-related corrections dominate your history, prompting you to convert a repeated correction into a permanent rule. ## Quick Start Ask the assistant to show my session stats and learning analytics dashboard.

Frequently Asked Questions about insights

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

FAQPage Schema
How do I see my coding session statistics and progress?

Ask for stats, analytics, or a dashboard and the Skill reads project memory files like LEARNED.md and learning-log.md plus git history. It then reports session duration, edits, corrections, learnings captured, and context usage.

How to track correction trends with an AI coding assistant?

The Skill counts [LEARN] entries and explicit correction markers in session history, then groups them by category in a heatmap. It flags corrections repeated many times but never captured as learnings so you can promote them to permanent rules.

What data sources does the session analytics report use?

It reads .claude/LEARNED.md or the LEARNED section of CLAUDE.md, .claude/learning-log.md, today's git log for the current user, and git diff stats. If files are missing, those sections are skipped gracefully.

Can it identify stale or unused learnings in project memory?

Yes, the report lists learnings that were captured but never applied, along with their age in days. It suggests removing entries that are no longer relevant to keep project memory clean.

Why does the insights report show no data for my session?

The report depends on existing project memory files and git history. If LEARNED.md, learning-log.md, or recent commits are absent, there is nothing to aggregate, so start capturing learnings during sessions first.