skills.rest
akoita
akoitaCOMMUNITY
Explore all skills by this author →
𝕏@aboobakarGitHub

setup-agent-toolkit

Inspect, preview, and install Agent Toolkit skills and policy files with conflict detection and backups.

Updated Jul 9, 2026
Search Tags:#claude code#codex#configuration#orchestration#backup#installation#agent toolkit
One-click install
npx skills add https://github.com/akoita/agent-toolkit --skill setup-agent-toolkit-akoita
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: setup-agent-toolkit
Source: https://github.com/akoita/agent-toolkit/tree/main/tools/setup-agent-toolkit
Command: npx skills add https://github.com/akoita/agent-toolkit --skill setup-agent-toolkit-akoita

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Installing agent orchestration skills and editing AGENTS.md or CLAUDE.md routing policy by hand risks overwriting existing developer configuration, breaking symlinks, or silently replacing user-owned files. This Skill enforces a read-only-first workflow with previews, approvals, backups, and rollback guidance. ## Core Features & Use Cases - Safe installation lifecycle: Inspect, preview, install, update, or repair Codex Maestro and Claude Maestro skills, workers, and agents at global or project scope. - Deterministic policy editor: The bundled configure_policy.py script previews a unified diff of the managed orchestration policy block and only writes with --apply, using atomic replacement, timestamped backups, and marker validation. - Conflict protection: Detects modified destinations, symlinks, malformed markers, and legacy worker files, refusing to overwrite anything without explicit user approval. - Use Case: A developer asks their agent to set up Codex Maestro globally; the Skill inspects existing installations, shows the exact AGENTS.md diff, waits for approval, installs the skill and worker definitions, then validates and reports rollback steps. ## Quick Start Ask the agent to set up Agent Toolkit for Codex at global scope and preview every change before applying it.

Frequently Asked Questions about setup-agent-toolkit

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

FAQPage Schema
How do I install Agent Toolkit skills for Codex or Claude Code?▼

Run the setup workflow which inspects existing destinations, previews changes, and installs only after approval. For Codex it uses the bundled install.py without --force; for Claude it copies the maestro skill and agent definitions to the selected skills and agents directories.

How do I add orchestration policy to AGENTS.md or CLAUDE.md safely?▼

Use the bundled configure_policy.py script with --platform and --scope to preview a unified diff of the managed policy block. It writes nothing until rerun with --apply, then creates a timestamped backup and replaces the file atomically.

Does the setup overwrite my existing agent configuration?▼

No. Existing destinations are compared against the source and reported as identical, modified, older, or symlinked. A modified destination is treated as a conflict and is never replaced without an explicit approved diff and a verified backup.

Can I install Agent Toolkit at project scope instead of globally?▼

Yes. Project scope copies the skill into .agents/skills/ or .claude/skills/ and worker definitions into .codex/agents/ or .claude/agents/ inside the repository. Global and project copies are never both installed unless explicitly requested.

What happens if the policy markers in my instruction file are malformed?▼

The policy editor refuses to proceed when it finds duplicate or mismatched managed markers, symlink targets, or non-UTF-8 content. It reports the problem and asks you to resolve it manually rather than guessing.

How do I roll back changes made to AGENTS.md or CLAUDE.md?▼

Restore the timestamped backup reported after the change, but only after previewing the reverse diff. If the file was newly created, remove just the managed block after confirming no other content was added.

Related Skills

Explore More Software Engineering Skills→
davila7davila7
30.5k

agent-management

Create, manage, and orchestrate AI agents through the AI Maestro CLI.

Community
Intermediate
pecorapromotionsllc-cellpecorapromotionsllc-cell

using-agent-skills

Discover and invoke the correct agent skill for a given task.

Community
Advanced
alter123-zzalter123-zz
18

skill-vetter

Detect security, permission, and malicious-code risks in third-party AI skills.

Community
Intermediate
videojsvideojs
931

maintain-agent-docs

Maintain AGENTS.md and CLAUDE.md guidance by removing duplication and enforcing context budgets.

Official
Intermediate