ai-native-cli
Design spec with 98 rules for building CLI tools that AI agents can safely use. Covers structured JSON output, error handling, input contracts, safety guardrails, exit codes, and agent self-description.
45
47%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/antigravity-awesome-skills-claude/skills/ai-native-cli/SKILL.mdQuality
Discovery
67%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 strong in specificity and distinctiveness, clearly carving out a unique niche around agent-friendly CLI tool design. However, it lacks an explicit 'Use when...' clause, which caps completeness, and could benefit from more natural trigger terms that users might actually say when seeking this guidance.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when designing or reviewing CLI tools intended for AI agent use, or when the user asks about machine-readable CLI output, agent-safe commands, or command-line tool design patterns.'
Include common user-facing trigger term variations such as 'command line', 'terminal commands', 'shell tools', 'machine-readable output', and 'agentic tooling'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete areas: structured JSON output, error handling, input contracts, safety guardrails, exit codes, and agent self-description. Also specifies the scope (98 rules) and purpose (building CLI tools for AI agents). | 3 / 3 |
Completeness | Clearly answers 'what' (design spec with 98 rules covering specific areas), but lacks an explicit 'Use when...' clause or equivalent trigger guidance. The 'when' is only implied by the domain description. | 2 / 3 |
Trigger Term Quality | Includes relevant terms like 'CLI tools', 'AI agents', 'JSON output', 'error handling', 'exit codes', but misses common user variations like 'command line', 'terminal', 'shell commands', 'machine-readable output', or 'agentic tooling'. | 2 / 3 |
Distinctiveness Conflict Risk | The niche is very specific: CLI tools designed for AI agent consumption, with safety guardrails and structured output. This is unlikely to conflict with general CLI skills, general coding skills, or general AI skills. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is a comprehensive reference document for the Agent-Friendly CLI spec, but it suffers from significant verbosity and poor progressive disclosure — all 98 rules are inlined into a single massive file. While the phased checklist and structured examples provide some actionability, the lack of executable implementation code and validation checkpoints limits practical utility. The content would benefit greatly from aggressive trimming of explanatory prose and splitting detailed rule listings into separate bundle files.
Suggestions
Split the detailed rule listings for each certification level into separate bundle files (e.g., level1-core.md, level2-recommended.md, level3-ecosystem.md) and keep only the overview and quick-start checklist in SKILL.md.
Remove sections that explain concepts Claude already knows (Common Pitfalls, Best Practices do/don't lists, the Overview paragraph restating what the skill does) to cut token usage by ~30-40%.
Add executable code examples showing how to implement key patterns — e.g., a Python/Node error handler function, a JSON output wrapper, or a guardrail validation function.
Add validation checkpoints to the implementation checklist — e.g., 'Run `mycli list | jq .` to verify O2 compliance' after Phase 1, or a compliance-checking script reference.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines, explaining concepts Claude already understands (what JSON output is, what exit codes mean, what stderr is for). The 'Core Philosophy', 'Overview', 'When to Use This Skill', 'Common Pitfalls', 'Best Practices', and 'Limitations' sections are largely redundant filler. The 98 rules could be presented far more compactly as a reference table rather than prose-heavy sections with extensive commentary. | 1 / 3 |
Actionability | The skill provides concrete rule lists, example JSON structures, and a phased checklist, which is useful. However, it lacks executable implementation code — there are no actual code snippets showing how to implement the error handler, JSON output wrapper, or guardrail checks. The bash examples show CLI invocation but not how to build the tool itself. | 2 / 3 |
Workflow Clarity | The three-phase implementation checklist provides a clear sequence, and the certification levels create a logical progression. However, there are no validation checkpoints — no way to verify compliance at each phase, no test commands to confirm rules are met, and no feedback loop for catching implementation errors during the build process. | 2 / 3 |
Progressive Disclosure | All 98 rules and three certification levels are crammed into a single monolithic file with no bundle files to offload detail. The full rule listings for Level 2 and Level 3 should be in separate reference files, with SKILL.md providing only the overview and quick-start checklist. The content is a wall of text that would benefit enormously from splitting into separate files per certification level. | 1 / 3 |
Total | 6 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- sickn33/antigravity-awesome-skills
- Commit
22710e9
- Reviewed
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.