Persistent local orchestrator for multi-task projects on the Hall of Automata. Plan, dispatch, and coordinate specialist agents via a design conversation with Old Major.
Exit Old Major session mode and clean up session files
List, view, or prune saved Tier-2 subagent consultation outputs
Dispatch ready tasks to the Hall as GitHub Issues with quota stewardship
Run preflight diagnostics for the Hall of Automata plugin environment
Provision the Hall of Automata Projects v2 board, custom fields, and labels on the current target repo — idempotent
Modifies files
Hook triggers on file write and edit operations
External network access
Connects to servers outside your machine
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Most AI coding tools work file by file, one change at a time. Hall of Automata works at project scale.
You open a session, describe what you want to build, and Old Major — an orchestrator running entirely inside Claude Code — designs the work, breaks it into tasks, and dispatches each task to a specialist AI agent. Every agent runs in its own autonomous Claude session, opens a pull request when done, and Old Major reviews and merges it. While agents are working, you can keep building. When they finish, Old Major syncs, unblocks the next wave, and keeps going.
This plugin brings that loop to completion: persistent session state, GitHub Issues as the coordination layer, automated review and merge, cross-invoker board sync, Google Drive context ingestion, and fully unattended operation. It is the bridge between a human team's intent and an AI team's execution — a methodology, not just a tool.
Dispatch mode requires a Hall of Automata instance — a GitHub org with the Hall of Automata GitHub App installed. Without one, Hall runs in local mode: Old Major plans and implements inline, inside your current Claude session, with no external agents.
Complete all four steps before installing.
Download and install Claude Code from claude.ai/code. It runs as a desktop app on Mac and Windows, and as an extension for VS Code and JetBrains IDEs.
The GitHub CLI (gh) handles authentication between Claude and GitHub.
gh auth login
It opens a browser window and walks you through signing in to GitHub.How to open a terminal:
- Mac: press
Cmd + Space, type "Terminal", press Enter- Windows: press
Win + R, typecmd, press Enter- Inside Claude Code or VS Code: press
Ctrl+`to open the built-in terminal
Hall uses GitHub's API to read and write issues, pull requests, and project boards on your behalf. This requires a Personal Access Token added to Claude's settings.
Step 1 — Create the token
Click Generate new token (classic)
Name it Hall of Automata CLI
Set an expiration (or choose No expiration)
Select exactly these scopes:
| Scope | Why it's needed |
|---|---|
✅ repo | Read and write repositories — issues, pull requests, branches, code |
✅ read:org | Read your organisation's team membership (used to verify invoker status) |
✅ project | Create and update GitHub Projects v2 boards and fields |
Click Generate token and copy it immediately — it is only shown once. It starts with ghp_.
This token has broad access to your repositories. Keep it private and do not share it.
Step 2 — Add it to Claude's settings
Locate Claude's global settings file on your computer:
| Operating system | File path |
|---|---|
| Mac | ~/.claude/settings.json |
| Windows | %APPDATA%\Claude\settings.json — paste this directly into the address bar in File Explorer |
| Linux | ~/.claude/settings.json |
Open the file in any text editor (Notepad works on Windows). Add an env section with your token. If the file already has content, add only the env block — don't replace what's there:
{
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
}
}
Save and restart Claude Code.
For dispatch mode, the Hall of Automata GitHub App must be installed on the GitHub organisation that owns your target repository. An org owner can install it from the Marketplace page in a few clicks.
Skip this step if you are using local mode only.
Inside any Claude Code session:
/plugin marketplace add MockaSort-Studio/marketplace
/plugin install hall-of-automata-cli@mockasort
Works in the desktop app, VS Code, JetBrains, and the CLI — no terminal or git required.
Hall is built around a session — an active working context that you open at the start of your work and close when you are done. Opening a session is intentional: it pulls personas, assembles the instruction stack, and starts the background watcher. It is not a fire-and-forget operation, and it is not meant to be opened and closed repeatedly.
The typical flow:
npx claudepluginhub mockasort-studio/marketplace --plugin hall-of-automata-cliPermanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Intelligent prompt optimization: injects the right context at the right moment so Claude lands a better first output. Clarifies vague prompts with research-based questions, plus targeted nudges for approach selection, plan readability, workflow routing, background execution, subagent routing, output readability, user-decision questions, and plan-mode assessment
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses