manage

Manage agents and skills in `.claude/` with atomic updates and permission changes.

24|3|Updated Feb 22, 2026
One-click install
npx skills add https://github.com/Borda/.home --skill manage-borda
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: manage
Source: https://github.com/Borda/.home/tree/main/.claude/skills/manage
Command: npx skills add https://github.com/Borda/.home --skill manage-borda

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the management of AI agents and skills within the .claude/ directory, ensuring consistency and preventing broken references during lifecycle operations.

Core Features & Use Cases

  • Lifecycle Management: Create, update (rename), and delete agents and skills.
  • Atomic Operations: Ensures updates and deletions are safe and prevent data loss.
  • Permission Management: Atomically add or remove permissions from settings.json and permissions-guide.md.
  • Use Case: When introducing a new specialized agent for security analysis, use this Skill to create it, update relevant documentation, and ensure its permissions are correctly configured without manual file edits.

Quick Start

Use the manage skill to create a new agent named 'code-reviewer' with a description of 'Reviews code for style and potential bugs'.

Frequently Asked Questions about manage

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

FAQPage Schema
How do I manage agent lifecycle operations in the .claude directory without breaking references?

Atomic agent lifecycle management in the .claude directory ensures safe creation, updates, and deletions while automatically propagating cross-references. This prevents broken links and data loss during renaming or removal operations.

What is the best way to atomically update permissions in settings.json and permissions-guide.md?

Atomic permission management updates settings.json and permissions-guide.md simultaneously to add or remove permissions. This ensures configuration consistency across both files without requiring manual edits.

How do I create a new agent with specific permissions using kebab-case naming?

Creating a new agent using kebab-case names involves executing specific commands for agent creation and permission configuration. The system atomically generates the agent files and configures associated permissions.

Can I rename an agent and automatically update all cross-references in my workflow configuration?

Renaming an agent triggers automatic cross-reference propagation throughout the .claude directory. The atomic update operation ensures all dependent files and configurations reflect the new name without manual intervention.

Why do manual updates to skill management files cause broken references?

Manual skill management updates bypass atomic operations, risking broken cross-references between agents, skills, and configuration files. Without atomic propagation, renaming or deleting files leaves dangling references that disrupt workflow automation.

Does this skill management approach support both agents and skills lifecycle operations?

Skill management supports both agents and skills through dedicated commands for creation, updates, and deletions. Both entity types use kebab-case naming and benefit from atomic cross-reference propagation.