tolibeartolibearCommunityยท1 Agent Skills Included

goalbuddy

Keeps AI coding agents on track during long goal runs

Gives Codex and Claude Code a structured operating loop for long, multi-step coding goals with a charter, task board, and proof receipts. Eliminates lost context, repeated planning, and unfinished work when a coding session pauses, resumes, or switches between AI tools. Coordinates Scout, Worker, and Judge sub-agents so every task is bounded, verified, and recorded until the goal is provably complete.
npx skills add tolibear/goalbuddy --all -g -y
Available:

Tells the AI agent how to run the goal loop as a project manager, assigning read-only Scout and Judge roles and one write-capable Worker per task.

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install GoalBuddy?โ–ผ

Run `npx skills add tolibear/goalbuddy --all -g -y` in your terminal to install the full suite globally for your AI coding agent.

What problem does GoalBuddy solve?โ–ผ

It keeps AI coding agents focused during long tasks by giving each run a clear goal, a live task board, and proof receipts so work can pause and resume without losing the plan.

Does GoalBuddy work with both Codex and Claude Code?โ–ผ

Yes. It installs natively into both tools, and a goal board started in one harness can be resumed in the other since everything lives as plain files in your repo.

How does GoalBuddy verify work is actually done?โ–ผ

Every task must produce a receipt with verification results, and a final Judge or PM audit maps all receipts back to the original goal before marking it complete.

Do I need to manage the task board myself?โ–ผ

No. The main agent thread acts as project manager, activating one task at a time and delegating work to Scout, Worker, and Judge sub-agents automatically.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’