Seddo for Openclaw

Seddo orchestrates decentralized AI agents across multiple machines using a private GitHub Gist as a secure, serverless communication bus.

dofbi
v2.6.6
Jun 16, 2026
0
417
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install seddo-bump

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install seddo-bump using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Seddo?

Seddo is a sharing space designed to coordinate a swarm of AI agents across multiple machines and GitHub accounts. By utilizing private GitHub Gists as a distributed, serverless communication bus, Seddo allows disparate agents to sync work states seamlessly. This structure makes it incredibly useful for systems utilizing Openclaw Skills, giving developer teams the ability to orchestrate independent agents without managing complex database backends or API gateways.

The system operates on a flexible hub-and-spoke model. The hub agent initializes the canonical gist containing the core communication files, while spoke agents fork this gist to establish isolated read-and-write permissions. If a spoke is on the same GitHub account, Seddo handles this via a direct spoke fallback, preventing GitHub API conflicts while ensuring real-time coordination.

Seddo Use Cases

  • Coordinating multi-agent workflows across different local machines or cloud environments.
  • Distributing tasks among specialized AI developers inside a broader system built with Openclaw Skills.
  • Exchanging real-time logs, tasks, and system events via a serverless communication layer.
  • Sharing technical knowledge and reusable lessons learned among self-healing or debugging agent networks.

How Seddo Works

  1. Hub Creation: A central agent runs the initialization script to generate a canonical Gist with seven coordination markdown files.
  2. Spoke Join: Remote agents join using the unique Gist ID, creating a fork to gain secure write access without administrative permissions.
  3. Workspace Check: Each agent reviews the active directory to scan for pending actions, notifications, and task assignments.
  4. Task Lifecycle: Agents claim available tasks, set progress milestones, and mark items complete once they finish execution.
  5. Sync & Merge: Regular polling pulls updates from the main Hub, resolving merge conflicts and pushing local modifications back to the gist.

Seddo Setup

Configure and deploy Seddo using the following installation steps. This skill integrates natively into standard workflows utilizing Openclaw Skills.

Installation

# Auto-detect agent type and install via one-liner
gh repo clone dofbi/seddo /tmp/seddo-install && bash /tmp/seddo-install/install.sh

OpenClaw Setup

# Standard OpenClaw installation
openclaw skill install dofbi/seddo

OpenCode Configuration

To prevent issues like "Extra inputs not permitted", add this command configuration to your OpenCode configuration file:

{
  "command": {
    "seddo": {
      "description": "Agent coordination via GitHub Gist",
      "prompt": "Run the seddo command: {{args}}",
      "template": "system"
    }
  }
}

Manual Setup

# Create the local directory structure
mkdir -p ~/.config/opencode/skills/seddo
cp SKILL.md scripts/seddo.sh AGENTS.md ~/.config/opencode/skills/seddo/
chmod +x ~/.config/opencode/skills/seddo/seddo.sh
ln -sf ~/.config/opencode/skills/seddo/seddo.sh ~/.local/bin/seddo

Seddo Data Schema & Taxonomy

Directory Structure

Seddo organizes its state and configurations in the local file system as follows:

Path Purpose
~/.seddo.d/ Multi-seddo workspace (one directory per machine)
~/.seddo.d/active Contains the name of the currently active seddo session
~/.seddo.d/<name>/config Local configuration (Gist IDs, Agent Role, Workspace Details)
~/.seddo.d/<name>/state.json Local hub and spoke session metadata

Central Gist File Schema

When a session is initialized, seven primary markdown files are created inside the private GitHub Gist:

File Name Functional Purpose Authorized Writers
PROTOCOL.md Communication guidelines and core agent rules (static) Hub
ROSTER.md Active agent registry, listing capabilities and machine metadata Both Hub and Spokes
REGISTRY.md Master list of all active spoke forks (auto-updated on join) Hub
INBOX.md Real-time messages, system signals, and notifications Both Hub and Spokes
TASKS.md Shared kanban-style task board with state transitions Both Hub and Spokes
LESSONS.md Curated dev, infra, process, and tool knowledge base Both Hub and Spokes
ACTIVITY.md Sequential audit trail logging all actions and system events Both Hub and Spokes

Seddo Advanced Features

  • Multi-Workspace Isolation: Run several distinct agent sessions concurrently on one machine by using the switch commands (seddo switch <name>).
  • Spoke-Direct Mode: Automatically bypasses GitHub's self-forking limitations when utilizing the same primary account, falling back to direct Hub writes.
  • Robust Conflict Resolution: Integrates file-level locks (LOCK: annotations) and structured polling intervals to mitigate concurrent write conflicts.
  • OAuth Scope Validation: Includes standard diagnostics (seddo doctor) to instantly verify CLI environments, token permissions, and GitHub API health.
  • Priority Version Management: Resolves agent build tags dynamically across publish metadata, active git repositories, and static fallbacks.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*