kingbootoshi

kingbootoshi / goal-ledger

Public

Single-file agent goal ledger for Claude Code and Codex CLI - pairs with /goal mode

12
0
89% credibility
Found May 20, 2026 at 11 stars 6x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Goal Ledger is a productivity skill for AI coding assistants that prevents work from getting lost during long, multi-session projects. When you start a goal with your AI, it creates a single readable webpage that tracks everything: what was done, what's in progress, what decisions were made, and what comes next. If your AI session gets interrupted or handed off to another session, the next AI simply reads this page and picks up exactly where it left off. It works with both Claude Code and OpenAI Codex CLI, and the whole system is designed so nothing falls through the cracks during complex projects.

How It Works

1
🤖 You're working on something big

You have a complex project that will take many sessions with your AI coding assistant.

2
😰 The context keeps getting lost

Halfway through a long task, your AI forgets what it already did. You have to re-explain everything.

3
📦 You install Goal Ledger

One simple setup connects the skill to your AI assistant, and it starts working automatically.

4
🎯 You start a goal

You tell your AI what you want to accomplish, and Goal Ledger creates a fresh tracking page for it.

5
Your AI keeps working
âš¡
Session stays connected

Everything flows smoothly and your AI finishes the work.

💤
Session gets interrupted

No problem — the next AI picks up the page and knows exactly where to continue.

✅ Your project completes smoothly

Even across many sessions and hand-offs, your work stays on track and nothing gets lost.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is goal-ledger?

Goal-ledger is a skill plugin that gives AI coding agents a persistent memory. When you invoke `/goal` mode in Claude Code or OpenAI Codex CLI, this tool creates a structured ledger that tracks every checkpoint, decision, blocker, and validation result in a single HTML file you can open in any browser. The agent writes progress directly into `implementation-notes.html`, which survives conversation compaction and session handoffs. It works across multiple agents using the open Agent Skills standard, so you can use the same setup in Claude Code, Codex CLI, and other compatible tools.

Why is it gaining traction?

The core insight is that agents forget. They compact conversations, get interrupted, or hand off to new sessions and lose context. Goal-ledger solves this by embedding the ledger path directly into the active goal itself, so the next agent on the thread knows exactly where to read state. The HTML file serves double duty: it's human-readable for you and machine-parseable for the agent. The inline progress timeline means no separate JSON file to keep in sync. It also includes an explicit escape hatch mechanism that lets the agent honestly report blocked or incomplete work instead of faking completion.

Who should use this?

Developers running multi-hour or multi-day coding tasks with Claude Code or Codex CLI. If you've ever watched an agent re-explore the same codebase twice, this fills that gap. Teams doing iterative refactors, large migrations, or exploratory prototyping where context survival across sessions matters. Not useful if your tasks complete in a single message.

Verdict

Worth installing if you live in `/goal` mode. The credibility score sits at roughly 0.9%, which reflects the project's early stage and modest star count, but the documentation is thorough and the single-file Python script is easy to audit. It's a small, focused tool that solves a real problem for agent-heavy workflows.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.