retro

Automating end-of-session retrospective documentation captures completed work, ongoing progress, blockers, and decisions by summarizing session history and commit logs, then persisting updates directly to MEMORY.md and backlog statuses to reduce context drift.

Updating backlog status and session memory after development work uses session history and commit logs to classify items into completed, in-progress, or new discoveries, persisting those updates to MEMORY.md for an auditable session-close workflow.

A session retrospective classifies completed work versus new discoveries by analyzing session history and commit logs against existing backlog data, categorizing items into completed, in-progress, or newly identified findings before writing updates to MEMORY.md.

Running a retrospective workflow that updates MEMORY.md requires no external dependencies, operating within lightweight product development workflows to capture session outcomes and update backlog statuses directly.

You can use this retrospective skill for lightweight product development workflows to capture completed items, identify progressed items and blockers, and persist decisions to MEMORY.md as a repeatable session-close process.

You should not use an automated retrospective to update your backlog if your session lacks commit logs or backlog data, as the classification of completed, in-progress, and new discoveries relies on that session history to function accurately.