tune

Edit LifeOS system prompts on a feature branch with commits and pushes.

20|4|Updated Jan 7, 2026
One-click install
npx skills add https://github.com/nbramia/LifeOS --skill tune-nbramia
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tune
Source: https://github.com/nbramia/LifeOS/tree/main/.agents/skills/tune
Command: npx skills add https://github.com/nbramia/LifeOS --skill tune-nbramia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Tune LifeOS orchestrator behavior based on feedback about a bad response. Use when the user describes a problem with how LifeOS answered a query — wrong sources, too verbose, not persistent enough, poor tool selection, or missing details. It also covers Codex task execution issues (scope, persistence, and notification style). It makes a surgical prompt edit, branches, commits, and pushes to implement changes.

Core Features & Use Cases

  • Surgical prompt edits to the system prompt to improve source usage, tool selection, and response quality.
  • Versioned tuning workflow: create a branch, commit changes, and push to review.
  • Applies to LifeOS orchestrator and Codex task execution scenarios, with clear rollback paths.

Quick Start

Provide a concrete feedback example to trigger a surgical system-prompt edit and commit the change on a feature branch.

Frequently Asked Questions about tune

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

FAQPage Schema
How do I fix a LifeOS orchestrator giving bad responses with wrong sources?

To fix a LifeOS orchestrator giving bad responses with wrong sources, you apply a surgical prompt edit to the system prompt. This workflow creates a feature branch, commits the targeted prompt changes, and pushes them to improve subsequent response quality.

How do I adjust system prompts when Codex task execution is too verbose?

To adjust system prompts when Codex task execution is too verbose, you use a surgical prompt edit workflow. It branches the changes, commits the reduced verbosity directives, and pushes to update the orchestrator behavior for future tasks.

What is the best way to correct poor tool selection in LifeOS orchestrators?

The best way to correct poor tool selection in LifeOS orchestrators is through surgical system-prompt tuning. You provide feedback about the bad response, which triggers a prompt edit on a feature branch to enforce better tool selection and persistence.

Does tuning LifeOS prompts require manual version control?

Tuning LifeOS prompts does not require manual version control, as the workflow automatically creates a feature branch, commits the surgical prompt edit, and pushes the changes. This provides a clear rollback path for orchestrator behavior updates.

Why does my LifeOS orchestrator lack persistence across Codex tasks?

Your LifeOS orchestrator lacks persistence across Codex tasks due to insufficient directives in the system prompt. You can fix this by providing feedback about the missing persistence to trigger a versioned surgical prompt edit.