AI Native CLI

AI-Native CLI design spec. Use when building CLI tools, designing command-line interfaces, or scaffolding new CLI projects. Covers structured JSON output, error handling, input contracts, safety guardrails, exit codes, and agent self-description. Includes an audit protocol for verifying CLI compliance.

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add ChaosRealmsAI/agent-cli-spec --skill ai-native-cli
Manual install
git clone --depth 1 https://github.com/ChaosRealmsAI/agent-cli-spec.git /tmp/agent-cli-spec
mkdir -p ~/.claude/skills
cp -r /tmp/agent-cli-spec/ai-native-cli ~/.claude/skills/ai-native-cli

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Agent-Friendly CLI Spec v0.1

When building or modifying CLI tools, follow these rules to make them safe and
reliable for AI agents to use.

## Core Philosophy

1. **Agent-only** -- output is always JSON, no human-friendly mode
2. **Agent is untrusted** -- validate all input at the same level as a public API
3. **Fail-Closed** -- when validation logic itself errors, deny by default
4. **Verifiable** -- every rule is written so it can be automatically checked

## Layer Model

This spec now uses two orthogonal axes:

- **Layer** answers rollout scope: `core`, `recommended`, `ecosystem`
- **Priority** answers severity: `P0`, `P1`, `P2`

Use layers for migration and certification:

- **core** -- execution contract: JSON, errors, exit codes, stdout/stderr, safety
- **recommended** -- better machine UX: self-description, explicit modes, richer schemas
- **ecosystem** -- agent-native integration: `agent/`, `skills`, `feedback`, inline context

Certification maps to layers:

- **Agent-Friendly** -- all `core` rules pass
- **Agent-Ready** -- all `core` + `recommended` rules pass
- **Agent-Native** -- all layers pass

## Output Mode

Output is always JSON. No human-friendly mode. No flags needed.

```bash
$ mycli list              # JSON output, always
```

- All output is JSON to stdout. Period.
- No `--human`, no `--agent`, no mode switching.
- If a human needs to read it, pipe through `jq`.

## agent/ Directory Convention
Read the full skill on GitHub

Details

Skill name
ai-native-cli
Path
ai-native-cli/SKILL.md
License
MIT
Repo updated
15 March 2026

FAQ

What does the AI Native CLI skill do?

AI-Native CLI design spec. Use when building CLI tools, designing command-line interfaces, or scaffolding new CLI projects. Covers structured JSON output, error handling, input contracts, safety guardrails, exit codes, and agent self-description. Includes an audit protocol for verifying CLI compliance.

How do I install the AI Native CLI skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add ChaosRealmsAI/agent-cli-spec --skill ai-native-cli. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is AI Native CLI popular?

It lives in ChaosRealmsAI/agent-cli-spec, which has 3 GitHub stars and 0 forks, and is included in 1 curated skill list.

Related skills

Connect

9

redpanda-data/skills

Teaches how to build streaming data pipelines with Redpanda Connect (formerly Benthos): declarative YAML configs with the input/pipeline/output model, processors, and Bloblang mappings for transforming messages. Use when creating or editing a Connect pipeline config, wiring inputs and outputs (Kafka topics, S3, HTTP, databases), writing or debugging Bloblang, or running pipelines with `rpk connect run`. For diagnosing a failing pipeline (lint errors, metrics, tracing, dry runs), see `/redpanda:connect-debugging`; for change-data-capture connectors, see the per-database `/redpanda:connect-cdc-*

Software Development

Warden

10

thatmike1/claude-skills

All-day accountability on request. Build a flexible day plan together, keep the current block visible, and check in through T3 Code with desktop reminders.

Software Development

Connect Debugging

9

redpanda-data/skills

Diagnoses, validates, and monitors Redpanda Connect pipelines. Use when a Redpanda Connect pipeline fails, stalls, drops messages, or won't start; when linting a config with rpk connect lint; tuning log level or format; reading Connect metrics (Prometheus, statsd, json_api) or traces (OpenTelemetry); checking /ready and /ping health endpoints; running a dry-run connection test; or diagnosing connector failures including auth errors, TLS errors, Enterprise license errors, backpressure, and checkpoint/cache issues. Covers Redpanda Connect enterprise features and their config keys: enterprise con

Software Development

Portpilot Assistant

9

lingengyuan/my-skills

Inspect, diagnose, allocate, and free local development ports using the bundled PortPilot CLI. Use when finding available ports, inspecting port listeners, resolving port conflicts, freeing ports, or managing PortPilot configuration.

Software Development

Cue Research CLI

8

sensedeal/cue-skills

Cue 深度研究 CLI(@cueai/cue-cli,命令 cue)使用说明。用于行业/公司/人物/政策/合规类深度调研:一句话发起多智能体研究,输出带来源的 Markdown 报告,可落盘、可续接会话追问、可指定搭子模板。涉及「深度调研」「尽调」「行研」「公司研究」「竞品对比」「政策跟踪」「人物核查」等需要多源检索与交叉验证的场景时使用。

Software DevelopmentIn 1 list

Pele

9

lingengyuan/my-skills

Render Mermaid diagrams as SVG with Pele. Use its JavaScript API for custom rendering, theming and browser integration.

Software Development

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.