startup

Initialize a session after degraded BIOS boot with health checks and boot status reporting.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/rudi193-cmd/Aionic-Claude-Skills --skill startup-rudi193-cmd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: startup
Source: https://github.com/rudi193-cmd/Aionic-Claude-Skills/tree/main/skills/startup
Command: npx skills add https://github.com/rudi193-cmd/Aionic-Claude-Skills --skill startup-rudi193-cmd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

BIOS boot is degraded and subsystems are not reporting; this skill provides a manual bootstrap by performing the actions usually done by bridge-open.py.

Core Features & Use Cases

  • Health check: verify Postgres via Unix socket (psql -U sean -d willow)
  • Read prior handoff: query ganesha.handoffs for the most recent entry and resolve pointer to arm DB for full content
  • Pre-warm cache: load BASE 17 compact index to /tmp
  • Verify MCP: call mcp__willow__willow_status to confirm server responding
  • Verify fleet: call mcp__willow__willow_chat with a ping to test inference path
  • Report boot status: identify which subsystems are up or degraded
  • Load open work: read ganesha.open_work for current task list
  • Load feedback: read opus.feedback for behavioral corrections

Quick Start

Invoke the startup protocol when BIOS boot is degraded to manually perform the bridge-open steps and initialize a session.

Frequently Asked Questions about startup

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

FAQPage Schema
How do I manually recover a session when BIOS boot is degraded and subsystems are not reporting?

Manual boot recovery initializes a session by performing health checks, reading prior handoffs, pre-warming the cache, and verifying the MCP server and inference fleet to establish a stable startup when subsystems are not reporting.

What steps are required to verify MCP status and fleet health during a degraded boot?

Verifying MCP status involves calling mcp__willow__willow_status to confirm the server responds, and verifying the fleet requires calling mcp__willow__willow_chat with a ping to test the inference path during the manual startup process.

How does pre-warming the cache work during manual session startup?

Pre-warming the cache works by loading the BASE 17 compact index into the /tmp directory during manual session startup, ensuring faster access to data and stabilizing the system before completing the boot recovery.

When do I need to manually load open work and feedback during boot recovery?

You need to manually load open work and feedback when the automatic handoff did not load, requiring you to read ganesha.open_work for current tasks and opus.feedback for behavioral corrections to restore session context.

How do I read prior handoff data if the automatic boot handoff did not load?

Reading prior handoff data requires querying the ganesha.handoffs table for the most recent entry, then resolving the pointer to the arm DB to retrieve the full content needed to resume the session.

Does manual boot recovery require a Postgres health check before initializing the session?

Yes, manual boot recovery requires a Postgres health check via the Unix socket using psql -U sean -d willow to verify database connectivity before proceeding with cache pre-warming and fleet verification.