AsWali

CallBoard — AI skill for Claude Code

AI community

A shared to-do list for you and your AI coding agents.

How to install CallBoard

This entry records only its repository, not the path inside it, so there is no exact command to give. Open AsWali/CallBoard and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What CallBoard does

A shared to-do list for you and your AI coding agents.

Alternatives in AI

Memmy Agent — 🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memor 1k ★
Wax — Shared Single-file memory layer for all your agents, sub mili-second RAG over text, photo and video on Apple S 791 ★
Claudexor — Multi-harness control plane for Claude Code, Codex, Cursor, and OpenCode: quota-aware rotation across multiple 425 ★

README

Callboard

**A shared to-do list for you and your AI coding agents.**

When Claude Code or Codex works on your project, Callboard keeps its plan where you can see it: what's being done, by whom, and what's waiting on you. Agents bring you their questions with options and a recommendation, and list the things only you can do (an API key, an account, a deploy) with exact steps. You answer in one click, and the agent carries on.

The Callboard page for a made-up habit tracker called Pebble, updating live: Codex claims the settings screen task, a Claude Code session called streaks finishes the streak counter, and then a question under For you is answered with one click, which shows up on the task that was waiting on it.

Set it up using a coding agent by telling it this:

Read https://asror.dev/callboard-setup.md and set up Callboard for me

Why

See what your agents are doing. Stop scrolling back through a chat to find out what's done and what's next. It's all in one list, updated as they work.
They ask, you decide. Decisions that are yours come to you as questions, with the options and the agent's recommendation. If an agent can't wait, it goes ahead with its best guess, and redoes the work if you pick something else.
Know what's holding things up. "Waiting on you" shows what only you can do, starting with whatever frees the most work.
Several agents, no collisions. Claude Code and Codex, several sessions, several branches: each agent claims a task before starting it, so no two do the same work.
Just files in your project. Three Markdown checklists you can read on GitHub or edit by hand. No account, no server, no cloud.