cli-agent-evaluate

To evaluate a CLI tool against a specific agent challenge, provide the CLI command and the challenge index. The skill returns a structured evaluation envelope containing a 0–3 score, severity, check details, and an actionable workaround.

A CLI Agent Spec challenge evaluation is a targeted, isolated test that scores a CLI tool from 0 to 3 based on its performance against a single challenge, generating a machine-readable envelope with severity and workaround data.

Yes, you can run a single challenge check before a full multi-challenge evaluation. This skill is designed for fast, isolated assessments to quickly verify a specific behavior before committing to broader testing.

You need an environment where the CLI tool is directly accessible. The evaluation requires passing the challenge identifier and the CLI command to produce the structured score and workaround output.

The CLI evaluation output includes a structured envelope containing the Challenge identifier, Severity, Score from 0 to 3, Check details, and an Optional Workaround to guide agent-oriented fixes.

Use a single-challenge CLI evaluation when you need a fast, isolated check of a specific behavior or error. It is ideal for quick targeted assessments before running larger, multi-challenge evaluations.