session-startup

Establishes repository status and picks the next task from tracked data.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/matthabermehl/generation-b-kids-books --skill session-startup-matthabermehl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-startup
Source: https://github.com/matthabermehl/generation-b-kids-books/tree/main/.codex/skills/session-startup
Command: npx skills add https://github.com/matthabermehl/generation-b-kids-books --skill session-startup-matthabermehl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill quickly reconstructs repository state and verifies a clean baseline so work on long-running autonomous projects can begin safely and reproducibly, avoiding wasted effort when the environment is out-of-date or tests fail.

Core Features & Use Cases

  • Repository inspection: gathers recent commits, git status, and the latest progress log to orient the operator.
  • Baseline verification: runs the project's smoke check to ensure the baseline is green before making changes.
  • Task triage and desk view: lists failing tasks, selects exactly one next task, and writes a concise .agent/current_task.md containing goal, constraints, plan, evidence, and status.
  • Use Case: start a new development session after pulling changes or after repository compaction to safely pick and document the immediate next task.

Quick Start

Run the session-startup checklist by inspecting .agent/progress.log, running the baseline smoke script, listing failing tasks, and writing a concise .agent/current_task.md with the chosen task and required evidence.

Frequently Asked Questions about session-startup

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

FAQPage Schema
How do I safely start a long-running autonomous work session in a repository?

To start a safe work session, inspect recent git status and .agent/progress.log, run a baseline smoke test, list failing tasks, and write the selected task to .agent/current_task.md.

How do I triage failing tasks after pulling new changes to a repository?

Triage failing tasks after pulling changes by running scripts/agent/feature_list.py to list failures, selecting one next task, and documenting its goal, constraints, and baseline status in .agent/current_task.md.

Why should I verify a clean baseline before starting autonomous development?

Verifying a clean baseline ensures the environment is up-to-date and tests pass, avoiding wasted autonomous effort when the repository state is out-of-date or the smoke check fails.

What's the best way to reconstruct repository state for a new development chat?

Reconstruct repository state by gathering recent commits, reading .agent/progress.log, running the project's smoke script, and selecting exactly one failing task to establish a verified, compact desk view.

Do I need a progress log to use an automated session startup checklist?

A .agent/progress.log is required to orient the operator and reconstruct repository state, providing the necessary history to safely select the next failing task and write .agent/current_task.md.

When should I not use an automated task triage and baseline verification process?

Avoid automated task triage and baseline verification when starting a new chat immediately after repository compaction if the smoke script fails, as proceeding wastes effort on an unverified baseline.