cli-design

Design and build agent-first CLIs with HATEOAS JSON responses, context-protecting output, and self-documenting command trees. Use when creating new CLI tools, adding commands to existing CLIs such as joelclaw, or reviewing CLI design for agent-friendliness. Triggers on 'build a CLI', 'add a command', 'CLI design', 'agent-friendly output', or any task involving command-line tool creation.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable code and a clear command-creation workflow, but it is a monolithic single file that inlines large protocol/type-reference sections which would benefit from being split into one-level-deep reference files. Validation checkpoints for the build/install workflow are present but light.

Suggestions

Move the Streaming Protocol (ADR-0058) section and the full Response Envelope TypeScript type definitions into separate reference files (e.g. STREAMING.md, RESPONSE.md) and link to them from SKILL.md so the overview stays lean.

Add an explicit validation checkpoint to the "Adding a new command" workflow (e.g. verify the built binary and run a smoke command before installing to ~/.bun/bin/), and a validate→fix→retry loop for any ops/destructive next_actions like kubectl rollout restart.

Trim rationale prose that restates knowledge Claude already has (e.g. "Agents have finite context windows. CLI output must not blow them up.") to keep the body token-efficient.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable JSON/TS/bash examples that earn their place, but includes minor over-explanation of concepts Claude knows ("Agents have finite context windows", "Agents parse JSON; they don't parse prose"). Not a 5 because some rationale prose could be trimmed; not a 3 because the bulk is genuinely useful reference material.	4 / 5
Actionability	Provides fully executable, copy-paste-ready code: bash build/install commands ("bun build src/cli.ts --compile --outfile joelclaw"), full TypeScript envelope shapes, jq pipe examples, and Effect CLI command definitions covering the common cases.	5 / 5
Workflow Clarity	"Adding a new command" gives a clear 7-step sequence plus a checklist, but the rebuild/install steps lack an explicit validate-before-ship checkpoint and ops actions (kubectl rollout restart) lack a validate→fix→retry feedback loop. Not a 5 due to these minor validation gaps; not a 3 because most checkpoints are present.	4 / 5
Progressive Disclosure	Well-organized with clear section headers, but everything is inlined into one ~440-line SKILL.md with no reference files — the Streaming Protocol (ADR-0058) and full Response Envelope type defs clearly belong in separate reference files. Bundle contains only logo assets, which are not used for disclosure. Not a 4 because content that should be separate is inline; not a 2 because section headers provide real navigable structure.	3 / 5
	Total	16 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it states concrete capabilities, includes natural trigger phrases, and explicitly covers both what and when. It is concise yet comprehensive with no fluff.

Dimension	Reasoning	Score
Specificity	"Design and build agent-first CLIs with HATEOAS JSON responses, context-protecting output, and self-documenting command trees" lists multiple specific concrete actions (HATEOAS responses, context-protecting output, command trees) with comprehensive coverage, matching the anchor-5 example.	5 / 5
Completeness	It explicitly answers both what ("Design and build agent-first CLIs with...") and when ("Use when creating new CLI tools, adding commands to existing CLIs such as joelclaw, or reviewing CLI design"), matching the anchor-5 example with concrete trigger phrases.	5 / 5
Trigger Term Quality	Explicit "Triggers on 'build a CLI', 'add a command', 'CLI design', 'agent-friendly output'" provides comprehensive natural-term coverage including synonyms users would actually say.	5 / 5
Distinctiveness Conflict Risk	The niche (agent-first CLIs, joelclaw) is clear with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: joelhooks/joelclaw
Path: skills/cli-design/SKILL.md
Commit: 7094ced

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.