Skip to main content
The update-playbook skill helps you keep your Packmind playbook up to date by analyzing your coding sessions and proposing changes to your standards, commands, and skills. It bridges the gap between day-to-day development and playbook maintenance, so best practices stay current without manual effort.
Learn more about how skills work in the Skills Management section.

​
Prerequisites

  • The CLI must be installed: see CLI Installation
  • You must be authenticated: run packmind-cli login
  • You must have an existing Packmind organization with at least one package installed

​
When Does It Trigger?

The skill activates automatically when you use phrases like:
  • “update the playbook”
  • “sync standards”
  • “check if we need new standards”
It can also be triggered after coding sessions where new patterns, decisions, or conventions emerged.

​
How It Works

The skill follows a four-phase process to go from session context to actionable playbook changes.

​
Phase 1: Summarize the Session

The AI reviews your conversation and extracts key information:
  • What was built or changed
  • What patterns emerged or were established
  • What decisions were made and why
  • What mistakes were corrected

​
Phase 2: Analyze Existing Artifacts

Using the session summary, the AI evaluates your current playbook across three domains in parallel:
  • Standards — Are existing standards still accurate? Should new lintable rules be added?
  • Commands — Do any existing commands need updating based on recent work?
  • Skills — Are current skills still relevant? Should new capabilities be captured?

​
Phase 3: Propose a Change Report

The AI consolidates its analysis into a numbered change report covering:
  • Updates to existing skills, standards, or commands
  • New standards or skills to create
  • Artifacts to deprecate
You can then choose which changes to apply — all of them, a specific selection, or everything except certain items.

​
Phase 4: Apply Approved Changes

For each approved change, the AI:
  1. Edits the relevant local artifact files or delegates to the appropriate creation skill
  2. Runs packmind-cli diff to preview the changes
  3. Submits the changes as proposals for human review on Packmind
  4. Optionally propagates validated changes locally with packmind-cli install --recursive

​
Tips

  • Run it regularly — Trigger the skill after meaningful coding sessions to capture patterns while they are fresh
  • Review proposals carefully — The skill submits changes as proposals, so your team can review and approve them on Packmind before they are distributed
  • Combine with distribution — After approving proposals, distribute updated packages to keep all repositories in sync. See Distribute Your Package