What are the different report depths available?

The skill offers three modes: Quick (5-10 min) for small tasks, Standard (15-20 min) for typical sessions, and Comprehensive (30+ min) for deep-dive analysis of major changes.

What is a Claude Code retrospective?

A structured analysis of an AI-assisted coding session that examines what was accomplished, how the AI performed, and how the workflow can be improved for future tasks.

Where are the retrospective reports stored?

Validated retrospective reports are saved in the .claude/skills/retrospecting/reports/ directory within your local project folder.

How does the skill gather data?

It automatically retrieves and synthesizes information from git commit history, native Claude session logs, project file metrics, and direct user feedback.

Can this skill update my project settings?

Yes, it can suggest and apply improvements to CLAUDE.md and SKILL.md files based on insights and successful patterns identified during the retrospective.

Session Retrospective

Name: Session Retrospective
Author: bitwarden

bybitwarden

Analytics & Monitoring

Analyzes Claude Code sessions and git history to generate actionable retrospective reports for continuous workflow improvement.

About

The Session Retrospective skill provides a structured framework for evaluating AI-driven development sessions. By synthesizing data from git commits, session logs, and user feedback, it generates detailed reports that highlight successful patterns, identify technical debt, and suggest improvements to project configuration files. It is an essential tool for developers and teams looking to optimize their collaboration with Claude and maintain high code quality across complex development tasks.

Key Features

Multi-source data collection from git history, session logs, and sub-agent interactions
0 GitHub stars
Quantitative metric tracking for tool success rates and completion efficiency
Tiered analysis depths ranging from Quick summaries to Comprehensive deep-dives
Automated suggestions for improving CLAUDE.md and SKILL.md project configurations
Pattern recognition to identify successful approaches and recurring anti-patterns

Use Cases

Post-mortem analysis of bug-fixing sessions to prevent future regressions
Reviewing complex feature implementations to identify architectural improvements
Codifying successful workflow patterns into project instructions to optimize AI behavior

About

Key Features

Multi-source data collection from git history, session logs, and sub-agent interactions
0 GitHub stars
Quantitative metric tracking for tool success rates and completion efficiency
Tiered analysis depths ranging from Quick summaries to Comprehensive deep-dives
Automated suggestions for improving CLAUDE.md and SKILL.md project configurations
Pattern recognition to identify successful approaches and recurring anti-patterns

Use Cases

Post-mortem analysis of bug-fixing sessions to prevent future regressions
Reviewing complex feature implementations to identify architectural improvements
Codifying successful workflow patterns into project instructions to optimize AI behavior