skills.rest
mpsuesser
mpsuesserCOMMUNITY
Explore all skills by this author →
GitHub

axi

Standardize agent-facing CLI stdout with TOON output and structured errors.

Updated Mar 12, 2026
Search Tags:#error handling#cli#token efficiency#structured output#toon#session hooks#agent ux
One-click install
npx skills add https://github.com/mpsuesser/workspace --skill axi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: axi
Source: https://github.com/mpsuesser/workspace/tree/main/dotconfig/pi/agent/skills/__techs__/axi
Command: npx skills add https://github.com/mpsuesser/workspace --skill axi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AXI standardizes how agent-facing CLI tools should communicate so autonomous agents can reliably interpret outputs and complete next actions without wasting tokens.

Core Features & Use Cases

  • Token-efficient stdout: Uses TOON for structured, agent-readable output with reduced token cost compared to equivalent JSON.
  • Lean default schemas: Minimizes fields in list outputs and supports an explicit mechanism (like a --fields flag) to fetch more only when needed.
  • Practical truncation strategy: Provides truncated previews of large fields along with clear “how to get the full content” guidance.
  • Actionable aggregates and empty states: Includes total counts and definitive “nothing found” messaging to prevent unnecessary re-checks.
  • Agent-safe operational rules: Uses structured errors in stdout, avoids interactive prompts, supports idempotent mutations, and recommends session-start hooks for ambient context.

Quick Start

Instruct your AI assistant to “Review my CLI’s list, detail, and mutation commands against AXI and propose changes so stdout is TOON-based, token-efficient, truncates large fields responsibly, and returns structured errors with useful next commands.”

Frequently Asked Questions about axi

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

FAQPage Schema
How do I make CLI output token-efficient for autonomous agents?▼

To make CLI output token-efficient for autonomous agents, use TOON for structured stdout and lean default schemas. Provide truncated previews of large data and use a --fields flag to fetch more only when needed.

What is the best way to format CLI errors for AI agents?▼

The best way to format CLI errors for AI agents is to output structured errors in stdout with correct exit codes. Avoid interactive prompts and include useful next commands so the agent can continue the workflow autonomously.

How do I prevent autonomous agents from wasting tokens on large CLI outputs?▼

Prevent token waste by applying practical truncation strategies that provide previews of large fields alongside clear guidance on how to get the full content. Include total counts and definitive empty states to avoid unnecessary re-checks.

Does my CLI need non-interactive flags for shell-driven agent workflows?▼

Yes, your CLI needs non-interactive flag-only operations for shell-driven agent workflows. Agent-safe operational rules require avoiding interactive prompts and supporting idempotent mutations to ensure reliable execution.

How do I provide ambient context to an agent at session start?▼

Provide ambient context to an agent at session start by integrating optional session-start hooks. These hooks supply workspace environment context so the agent can execute shell-driven workflows reliably without extra queries.

Related Skills

Explore More Software Engineering Skills→
inbharataiinbharatai
22

cli-builder

Build Python CLI tools with argument parsing, prompts, and completions.

Community
Intermediate
duongdevduongdev
9

hook-optimization

Optimize CCPM Claude Code hooks for performance and token efficiency.

Community
Advanced
timurkhakhalevtimurkhakhalev
8

codex-cli-subagent

Spawn a nested Codex subagent via codex exec for non-interactive project tasks.

Community
Advanced
raykrayk
3

use-toon

Delegate agent tasks using TOON with schema.org vocabulary.

Community
Advanced