APM

>Agent Skill

@athola/catchup

skilldevelopment

Use this skill to quickly understand "what changed and what matters". Use when resuming work after absence, preparing handoff documentation, reviewing sprint progress, analyzing git history for context. Do not use when doing detailed diff analysis - use diff-analysis instead. DO NOT use when: full code review needed - use review-core instead.

apm::install
$apm install @athola/catchup
apm::skill.md
---
name: catchup
description: 'Use this skill to quickly understand "what changed and what matters".
  Use when resuming work after absence, preparing handoff documentation, reviewing
  sprint progress, analyzing git history for context. Do not use when doing detailed
  diff analysis - use diff-analysis instead. DO NOT use when: full code review needed
  - use review-core instead.'
category: analysis-methods
tags:
- summarization
- context-acquisition
- insights
- follow-ups
dependencies:
- imbue:proof-of-work
tools:
- git
- log-tools
usage_patterns:
- context-catchup
- handoff-preparation
- progress-review
complexity: intermediate
estimated_tokens: 700
progressive_loading: true
module_strategy: context-based
---
## Table of Contents

- [Overview](#overview)
- [When to Use](#when-to-use)
- [Activation](#activation)
- [Progressive Loading](#progressive-loading)
- [4-Step Methodology](#4-step-methodology)
- [Output Format](#output-format)

# Catchup Analysis Methodology

## Overview

Structured method for quickly understanding recent changes in git repositories, meeting notes, sprint progress, document revisions, or system logs. Answers "what changed and what matters?" efficiently.

## When To Use
- Joining ongoing work or returning after absence
- Before planning or reviewing handoffs
- Any "what happened and what's next" context

## When NOT To Use

- Doing detailed
  diff analysis - use diff-analysis instead
- Full code review needed
  - use review-core instead
- Doing detailed
  diff analysis - use diff-analysis instead
- Full code review needed
  - use review-core instead

## Activation
**Keywords**: catchup, summary, status, progress, context, handoff
**Cues**: "get me up to speed", "current status", "summarize progress"

## Progressive Loading

Load modules based on context:

**Git**: Load `modules/git-catchup-patterns.md` for git commands. Consider `sanctum:git-workspace-review` for initial data gathering.

**Documents/Notes**: Load `modules/document-analysis-patterns.md` for meeting notes, sprint tracking, document revisions.

**Logs/Events**: Load `modules/log-analysis-patterns.md` for time-series and metric analysis.

**Always Available**: `imbue:proof-of-work`, TodoWrite workflow, structured output.

## Required TodoWrite Items
1. `catchup:context-confirmed` - Boundaries established
2. `catchup:delta-captured` - Changes enumerated
3. `catchup:insights-extracted` - Themes identified
4. `catchup:followups-recorded` - Actions captured

## 4-Step Methodology

### Step 1: Confirm Context
Define scope (git branch, sprint, meetings), baseline (last state), and current target. See modules for commands.

### Step 2: Capture Delta
Enumerate changed items with metrics. Prioritize source/config/docs over generated artifacts. See modules for strategies.

### Step 3: Extract Insights
Per item: **What** (change), **Why** (motivation), **Implications** (tests/risks/deps). Rollup into themes.

### Step 4: Record Follow-ups
Capture: Tests, Documentation, Reviews, Blockers, Questions. If none, state explicitly.

## Output Format
```
## Summary
[2-3 sentence theme + risk overview]

## Key Changes
- [Item]: [what/why/implication]

## Follow-ups
- [ ] [Action with owner]

## Blockers/Questions
- [Item requiring resolution]
```
**Verification:** Run the command with `--help` flag to verify availability.

## Integration
Use `imbue:diff-analysis` for risk assessment, `imbue:proof-of-work` for reproducibility, `sanctum:git-workspace-review` for git data. Feed to `brainstorming` or `writing-plans` as needed.

## Token Conservation
Reference paths + lines (don't reproduce). Summarize outputs. Defer deep analysis. Use progressive loading.

## Exit Criteria
- Four TodoWrite items completed
- Context/delta/insights/follow-ups captured
- Stakeholders understand state without re-reading sources
## Troubleshooting

### Common Issues

**Command not found**
Ensure all dependencies are installed and in PATH

**Permission errors**
Check file permissions and run with appropriate privileges

**Unexpected behavior**
Enable verbose logging with `--verbose` flag