interview-me

github:addyosmani/agent-skills · first seen 2026-10-30 · source · badge

Extracts what the user actually wants instead of what they think they should want. Achieves this through one-question-at-a-time interview until ~95% confidence about the underlying intent. Use when an ask is underspecified ("build me X" without "for whom" or "why now"), when the user explicitly invokes ("interview me", "grill me", "are we sure?", "stress-test my thinking"), or when you catch yours

Works with

idea-refine
Shares stress, thinking, refine, idea0.25
grilling
Shares grill, stress, thinking, question0.23
loop-me
Shares grill, loop, interview, question0.18
using-agent-skills
Shares invokes, development, interview, requirements0.18
grill-me
Shares interview, grill, plan, description0.16

Used with

idea-refine
Shares stress, thinking, refine, idea0.25
grilling
Shares grill, stress, thinking, question0.23
loop-me
Shares grill, loop, interview, question0.18
using-agent-skills
Shares invokes, development, interview, requirements0.18
grill-me
Shares interview, grill, plan, description0.16
to-questionnaire
Shares question, interview, grill, catch0.16
constraint-driven-development
Shares constraint, development, down, driven0.15
doubt-driven-development
Shares stress, doubt, speed, development0.15
spec-driven-development
Shares guessing, requirements, development, spec0.13

Skill Health 100 (freshness and quality) estimated by backskills

Freshness 100 · Quality 100 · Efficacy not yet tested

Efficacy is shown once a skill has at least 10 use cases or a tools-enabled run. A small preliminary baseline has run and is not counted in this score.

Authors: run your own use cases with POST /api/skill-check (author-verified; never changes this public estimate). Paying never affects ranking. Author tests never change the public score. The public health score only breaks ties between equal matches.

Version history

2026-10-30T20:36:17 · sha256 ace4b675999b