What problem does it solve?
Setting up a self-hosted Cyrus agent requires many manual steps: installing prerequisites, configuring Claude authentication, exposing a webhook endpoint, creating Linear/GitHub/GitLab/Slack apps, and connecting repositories. This Skill orchestrates the entire onboarding flow so you can get Cyrus running as a background agent without following a lengthy manual guide.
Core Features & Use Cases
- Orchestrated Sub-Skills: Runs nine ordered sub-skills (prerequisites, Claude auth, endpoint, Linear, GitHub, GitLab, Slack, repository, launch) with skip-if-done logic so re-runs are safe.
- Browser Automation: Uses claude-in-chrome MCP tools or the agent-browser CLI to navigate web UIs, fill forms, and scrape credentials, with a manual guided fallback.
- Secret-Safe Configuration: Never reads or writes ~/.cyrus/.env directly; secrets are handled only through user-approved shell commands.
- Use Case: A developer wants Cyrus to monitor Linear issues and open GitHub PRs. They run /cyrus-setup, answer a few questions about agent name, surfaces, and package manager, and the agent completes the full installation and launch.
Quick Start
Ask your AI coding agent to run the cyrus-setup skill to install and configure Cyrus end-to-end for your chosen platforms.