What this pack does
The customer research pack turns your AI agent into a research partner. Build data-backed ICPs with real pain points and buying triggers. Run Voice of Customer analysis that captures the exact language your buyers use - not your interpretation, their words. Then build AI twin personas and test your messaging against a realistic buyer panel before you spend a dollar on ads.
The three skills feed into each other: /icp builds the profile, /voc captures how that persona talks, and /twins creates testable personas from both. Run them in sequence for a full research loop.
Skills
/icp Build an Ideal Customer Profile backed by real data. Researches pain points, buying triggers, objections, and where your buyers hang out. Cites sources - marks anything unvalidated.
/voc Voice of Customer analysis from Reddit, LinkedIn, forums, review sites, and podcasts. Extracts themes with frequency, emotional intensity, and exact quotes. Cross-validates across ICPs.
/twins Build a panel of 10 AI twin personas from an ICP, then test messaging against them. Each twin has specific pressures, decision style, and blind spots. Returns scores and recommended changes.