session-end

Capture lessons, update plans, and archive completion evidence at session end.

5|2|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/dbc-oduffy/coordinator-claude --skill session-end-dbc-oduffy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-end
Source: https://github.com/dbc-oduffy/coordinator-claude/tree/main/plugins/coordinator/skills/session-end
Command: npx skills add https://github.com/dbc-oduffy/coordinator-claude --skill session-end-dbc-oduffy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents completed work from leaving behind stale documentation, missing lessons, or unarchived outcomes so future sessions start with accurate context.

Core Features & Use Cases

  • Lessons Capture: Records durable engineering lessons into tasks/lessons.md when new, reusable insights were learned.
  • Plan and Status Updates: Updates feature-scoped plans, checks completed items, and refreshes status and review notes.
  • Completion Archiving: Creates archive/completed/YYYY-MM entries for substantive work that isn’t already tracked.
  • Optional Cross-Repo Memo Hygiene: Prompts for and actiones any resolved receiver-side cross-repo inbox memos in this repo.
  • Orientation Refresh via Pinboard: Writes a single pinboard line to tasks/orientation_cache.md only when the next session must see a critical transient gotcha.

Quick Start

Invoke the session-end skill to wrap up your completed work by capturing lessons and archiving what shipped, without doing any handoff.

Frequently Asked Questions about session-end

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

FAQPage Schema
How do I close a completed workstream and capture engineering lessons learned?

Closing a completed workstream involves capturing reusable insights into lessons documentation, updating plan status, and archiving completion evidence. This process ensures future sessions start with accurate context by preventing stale documentation and missing outcomes.

What is the best way to archive completed work and update plan documentation?

Archiving completed work requires creating dated archive entries for substantive tasks while refreshing feature-scoped plan status and review notes. This deterministic file operation approach records durable evidence without conflicting with concurrent agent operations across the repository.

Does session-end support concurrent agents working in the same repository?

Concurrent agent operations are supported by enforcing deterministic file operations and fan-in commit behavior. This approach prevents repo-wide conflicting operations, ensuring multiple agents can wrap up sessions independently without overwriting shared documentation or archive entries.

When should I use session termination instead of a handoff process?

Session termination applies when a workstream is fully finished and requires capturing lessons and archiving outcomes, rather than transferring active context. This exclusivity ensures completed work is properly closed out instead of leaving stale plans or unarchived results for future sessions.

Can I update orientation cache and resolve cross-repo memos during session wrap-up?

Orientation cache updates and cross-repo memo resolution are supported during wrap-up. A single pinboard line is written to cache critical transient gotchas, while optional memo hygiene prompts action on any resolved receiver-side inbox memos in the current repository.

Why does session-end create archive entries in YYYY-MM format?

Archive entries use YYYY-MM format to organize substantive completed work chronologically within the archive directory. This deterministic structure tracks shipped outcomes by month, ensuring completion evidence is preserved and easily retrievable for future session context.