Skip to content
SkillModel-led3,342 tokens in use

interview-me

Interview the user one question at a time, with a guess attached, until the underlying intent is clear and confirmed.

119,377 stars on the repository MIT Licenseaddyosmani/agent-skills · skills/interview-me
npx skills add addyosmani/agent-skills --skill interview-me
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing interview-me costs

108tokens idle
3,342tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure25/25
  • Examples15/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community76
  • Maturity32
  • Momentum40

Interview Me closes the gap between what people ask for and what they want, before any plan, spec or code exists. The agent writes a hypothesis with a confidence number, asks one focused question at a time with its own guess attached, and listens for answers that sound like convention or best practice.

When it can predict the user’s reaction to the next three questions, it restates the intent as outcome, user, why now, success, constraint and out of scope. It needs an explicit yes, then stops its turn and offers downstream paths.

What You Get

  • A six-step interview process
  • A restate template with an Out of scope line
  • A checkable 95 percent confidence stop test
  • A worked before and after example, red flags and a verification list

Common Use Cases

  • Clarifying a vague feature request
  • Surfacing hidden assumptions before planning
  • Choosing between competing values such as simplicity and flexibility
  • Handing a confirmed intent to the spec skill

What Makes It Unique

The stop condition is a test the agent can check: can it predict the user’s reaction to the next three questions. The skill also lists replies that do not count as a yes, such as whatever you think is best or sounds good, and it must stop its turn at confirmation.

Under The Hood

The skill is a single SKILL.md of instructions for conversation. It needs a live user and tells the agent not to run in CI or autonomous loops. It runs no scripts, and after confirmation it may offer to save the intent to docs/intent.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers