catchup

Aggregate repository state into a start-of-session briefing with alerts.

37|2|Updated Jun 28, 2025
One-click install
npx skills add https://github.com/jellyrock/jellyrock --skill catchup-jellyrock
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: catchup
Source: https://github.com/jellyrock/jellyrock/tree/main/.claude/skills/catchup
Command: npx skills add https://github.com/jellyrock/jellyrock --skill catchup-jellyrock

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill eliminates the tedious manual work of piecing together scattered repository state (git status, open PRs, CI runs, pending followups, upstream signal watchlists, and project journal staleness) at the start of a work session, ensuring you never miss urgent items or waste time reconstructing context after a gap.

Core Features & Use Cases

  • Single-call state aggregation: The built-in catchup-state.js script pulls all relevant repo state in one read, replacing dozens of manual git, GitHub, and file checks.
  • Deterministic banner alerts: Automatically surfaces urgent items like stale progress docs, failed CI runs, pending PR reviews, and outdated upstream signals without manual parsing.
  • Use Case: A developer resuming work on the JellyRock Roku client after a 3-day break can run this skill to instantly see if their branch has failing CI, if any PRs need their review, and what the highest-priority next task is.

Quick Start

Invoke the catchup skill at the start of any new work session to get a full repo state briefing with urgent alerts and a recommended next action.

Frequently Asked Questions about catchup

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

FAQPage Schema
How do I restore repository context and track open PRs when starting a development session?

To restore repository context at session start, you can aggregate scattered state like open PRs, CI run status, and pending followups into a single briefing. This eliminates manual context reconstruction by reading the project state cursor and watchlist to produce deterministic alerts.

What is the best way to monitor CI run status and project journal staleness across a codebase?

Monitoring CI run status and project journal staleness is best handled by a single-call state aggregation script. It pulls all relevant repo state in one read, replacing dozens of manual git and file checks to automatically surface urgent items like stale progress docs.

How do I get a start-of-session repo state briefing without manually checking git status and upstream signals?

Getting a start-of-session repo state briefing requires invoking a single aggregation command. The script reads the project's progress.md state cursor and signals-backlog.md watchlist, generating prioritized next-step suggestions and banner alerts without manual parsing.

Do I need specific files like progress.md or an ADR directory to track pending followups and repo state?

Yes, tracking pending followups and repo state requires access to the project's catchup-state.js aggregator script, progress.md state cursor, signals-backlog.md watchlist, and ADR directory to produce deterministic banner alerts and next-step suggestions.

Can I use a session-start briefing to track upstream signal watchlists for pending PR reviews?

Yes, a session-start briefing can track upstream signal watchlists and pending PR reviews. The aggregation script consolidates this scattered repository state, automatically surfacing outdated upstream signals and pending reviews as urgent deterministic banner alerts.

Why does manual context reconstruction fail when resuming work after time gaps in a development workflow?

Manual context reconstruction fails after time gaps because it requires piecing together scattered repository state like git status, open PRs, and CI runs. This tedious process risks missing urgent items, whereas a single-call state briefing aggregates all checks deterministically.