Abdulaziz-almoshen
User🛰️ Orbit — turn any repo into a self-prompting agentic loop (memory + sub-agent team + skills + loop + hard stop conditions). A Claude Code plugin.
Categories
Indexed Skills (2)
orbit-upgrade
Upgrade the installed orbit plugin to the latest version from GitHub and show what changed. Use when the user says "upgrade orbit", "update agentic loop", "get the latest orbit", or when the /orbit preamble reports UPGRADE_AVAILABLE. Speech-to-text aliases: "update agent loop", "upgrade the agent loop".
orbit
Transform any product repository into a production-grade, self-prompting agentic loop system following Daisy Hollman's "build a system that prompts itself" methodology. Audits the current project, then scaffolds persistent memory (CLAUDE.md + STATE files), a specialized sub-agent team, domain skills, a read→act→evaluate→update→decide loop, and HARD stop conditions (iteration/cost/ runtime caps, eval gates, human-approval checkpoints). Produces a model-agnostic core PLUS a Claude Code adapter, so the loop can run on your own orchestrator (e.g. Gemini) or via headless `claude -p`. Domain-agnostic: it characterizes whatever product it runs in and fits the system to it. Use this skill whenever the user wants to set up, upgrade, or "productionize" an autonomous agent / agentic loop / self-prompting system, build an orchestrator, add a sub-agent team or persistent memory, or mentions Daisy Hollman, "a system that prompts itself", "Ralph loop", agentic SDLC roles, or stop conditions for a runaway agent.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.