Organising Agent Skills

[!IMPORTANT] Mandatory User Confirmation Guardrail: The agent MUST NOT delete any skill directories, disable/enable skills in configuration files (skills.json), or mutate configuration/rules documents (GEMINI.md, AGENTS.md, README.md) without first presenting the audit findings, proposed actions, and token metrics to the user and receiving explicit confirmation to proceed.

This skill provides a repeatable, structured workflow for auditing, pruning, and organising agent skills in Antigravity and Gemini environments. It ensures your prompt context window remains lean (~30–50 active skills / ~3,000 tokens overhead per interaction turn) while retaining full access to all installed skills.

Table of Contents

• Overview & Token Budgeting
• Workflow Phase 1: Audit & Discovery
• Workflow Phase 2: Meta-Skill Delegation
• Workflow Phase 3: On-Demand Rules (GEMINI.md / AGENTS.md)
• Workflow Phase 4: Portability & skills.json Exclusions
• Workflow Phase 5: Documentation Update