first-run-setup

Bootstraps Node, Python, Ollama models, and MCP tooling for new orama-system machines.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/diazMelgarejo/orama-system --skill first-run-setup-diazmelgarejo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: first-run-setup
Source: https://github.com/diazMelgarejo/orama-system/tree/main/bin/orama-system/skills/first-run-setup
Command: npx skills add https://github.com/diazMelgarejo/orama-system --skill first-run-setup-diazmelgarejo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Setting up a new machine for the orama-system toolchain requires installing and validating many components—Node, Python 3.13, Ollama models, code-review-graph, gbrain, embeddings, and Claude Code profiles—which is error-prone and tedious to do manually. ## Core Features & Use Cases - Idempotent Bootstrap: Runs status checks and installs only missing components, safely resuming after interruptions without reinstalling satisfied binaries. - Cross-Platform Node Onboarding: Ships a dedicated Windows GPU node onboarding reference covering LM Studio requirements, Git Bash PATH shims, and LAN fleet coordination. - Error Resilience: Integrates the ADR-045 gstack/gbrain/CRG safety framework with retry-with-backoff and actionable error reporting. - Use Case: After cloning the repo on a fresh Windows RTX workstation, run the status and install actions to validate the toolchain, then confirm the ~/.orama-system/first-run.done marker before starting agentic work. ## Quick Start Ask the agent to run the first-run bootstrap status check and then install any missing orama-system toolchain components on this machine.

Frequently Asked Questions about first-run-setup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I bootstrap the orama-system toolchain on a new machine?

Run `bash bin/orama-system/scripts/first-run-install.sh status` to see what is missing, then run the same script with `install` to set up Node, Python 3.13, models, and related tooling. Success is confirmed by the `~/.orama-system/first-run.done` marker file.

How do I onboard a new Windows GPU node to the fleet?

Follow the references/windows-node-onboarding.md guide, which covers Git, NVM for Windows, Python 3.13, LM Studio as a hard requirement, and LAN coordination via a Task Scheduler pulse job. All commands are idempotent and safe to re-run after interruptions.

Is the first-run install script safe to re-run after a failure?

Yes, the install is idempotent and resumes from the last incomplete component rather than restarting from zero. The `--force` flag re-validates every component but only repairs config, never reinstalling satisfied binaries or models.

Why does first-run status show ollama as missing on Windows?

That is expected behavior, not a failure. Ollama is the hard requirement for Mac nodes, while Windows nodes hard-require LM Studio running on port 1234 instead, so the ollama line can be ignored on Windows.

Why does first-run-install.sh not find Python on Windows?

The script probes the literal binary name `python3.13`, which the Windows python.org installer does not create. Add an extensionless Bash shim named `python3.13` that execs python.exe, and put it on Git Bash's PATH.