CtrlK
BlogDocsLog inGet started
Tessl Logo

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 (joelclaw, slog), 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.

90

1.64x
Quality

88%

Does it follow best practices?

Impact

94%

1.64x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

100%

75%

Build a Developer Notes CLI

JSON envelope and HATEOAS next_actions

Criteria
Without context
With context

Effect CLI framework

0%

100%

Platform-node import

0%

100%

JSON-only output

100%

100%

ok field present

0%

100%

command field present

0%

100%

result field present

0%

100%

next_actions present

0%

100%

Template syntax in next_actions

0%

100%

params field for templates

0%

100%

Error envelope with fix

0%

100%

Command naming

60%

100%

Kebab-case flags

100%

100%

No --json flag

100%

100%

96%

16%

Extend the `devcli` Tool with a Cache Management Command

Self-documenting command tree

Criteria
Without context
With context

cache command defined

100%

100%

cache clear subcommand

100%

100%

cache inspect subcommand

100%

100%

withDescription on cache

0%

100%

withDescription on subcommands

0%

100%

Root commands array updated

100%

100%

cache added to subcommands

100%

100%

New commands return JSON envelope

100%

100%

Contextual next_actions

100%

100%

Existing commands preserved

100%

50%

No plain text output

100%

100%

88%

22%

Add an Event History Command to a Workflow CLI

Context-protecting output and truncation

Criteria
Without context
With context

events command defined

100%

100%

Default line limit

100%

100%

Truncation detection

0%

100%

full_output path when truncated

0%

0%

total count reported

100%

100%

withDescription on events

100%

100%

Root commands array updated

100%

100%

JSON envelope shape

100%

100%

next_actions with lines template

0%

100%

No unbounded output

100%

100%

Repository
joelhooks/joelclaw
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

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.