Claude MarketplacesSubmit
google-labs-code avatar

Agent Dx Cli Scale

google-labs-code/design.mdOfficial

This is a scoring rubric for evaluating how well any CLI is designed for AI agents rather than humans. It covers seven dimensions: machine-readable output, raw payload input, schema introspection, context window discipline, input hardening, safety rails, and agent knowledge packaging. Each axis scores 0 to 3, giving you a total out of 21. The scoring is opinionated and assumes agents fail differently than humans (they hallucinate resource IDs, not typos), so it rewards things like rejecting path traversals and percent-encoded segments. Use it when you're building or auditing a CLI that agents will actually call, not just wrap in a tool description.

skills.sh installs
166
GitHub stars
28.8K
last updated
Aug 21, 2026
View on GitHub

Install

$ npx -y skills add google-labs-code/design.md --skill agent-dx-cli-scale --agent claude-code

Installs into .claude/skills of the current project.

Files

Agent DX CLI Scale

Use this skill to evaluate any CLI against the principles of agent-first design. Score each axis from 0–3, then sum for a total between 0–21.

Human DX optimizes for discoverability and forgiveness. Agent DX optimizes for predictability and defense-in-depth. — You Need to Rewrite Your CLI for AI Agents

Scoring Axes

1. Machine-Readable Output

Can an agent parse the CLI's output without heuristics?

ScoreCriteria
0Human-only output (tables, color codes, prose). No structured format available.
1--output json or equivalent exists but is incomplete or inconsistent across commands.
2Consistent JSON output across all commands. Errors also return structured JSON.
3NDJSON streaming for paginated results. Structured output is the default in non-TTY (piped) contexts.

2. Raw Payload Input

SKILL.md on GitHub

Sponsors

Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthlyAgent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any language

Details

Repository
github.com/google-labs-code/design.md
Categories
AI & Agent Building, Design & UI/UX, CLI & Terminal
First seen
Jun 3, 2026

More from google-labs-code/design.md

All 4

Recommended

More AI & Agent Building
1qodex-ai avatardata-organization-systemqodex-ai/ai-agent-skillsOrganize and categorize files into logical structures. Creates folder hierarchies, renames files systematically, and consolidates related data.166skills.sh installs
2sickn33 avataragent-orchestratorsickn33/antigravity-awesome-skillsagent orchestrator166skills.sh installs
3sickn33 avataravoid-ai-writingsickn33/antigravity-awesome-skillsavoid ai writing165skills.sh installs
4wwwzhouhui avatarjimeng_mcp_skillwwwzhouhui/skills_collection使用jimeng-mcp-server进行AI图像和视频生成。当用户请求从文本生成图像、合成多张图片、从文本描述创建视频或为静态图像添加动画时使用此技能。支持四大核心能力：文生图、图像合成、文生视频、图生视频。需要jimeng-mcp-server在本地运行或通过SSE/HTTP访问。165skills.sh installs
5akillness avatarprompt-repetitionakillness/oh-my-skillsprompt repetition164skills.sh installs
6mindgames avataragent-browser-relaymindgames/agent-browser-relayRead metadata, DOM, screenshots, and structured page payloads from an attached Chrome or Chromium tab through the local Agent Browser Relay extension. Use when 164skills.sh installs