AI Agent Workflow Collection

Designs the first-run path that carries a brand-new user from signup to their aha-moment fast — defines the activation event, maps the current first-run, diagnoses the drop-off, and re-sequences the steps with triggers, nudges, and empty states. Use when a PM says "design our onboarding", "improve activation", "get users to the aha moment", "redesign first-run", "new-user flow", "why aren't new users sticking", "activation flow", or "shorten time-to-value". NOT for measuring activation/retention (cohort-analysis, metrics-review), mapping the whole journey (journey-map), or modeling whole-product loops (growth-loops).

Generate a Business Model Canvas with all 9 building blocks. Use when creating a business model, documenting how a business creates value, or analyzing an existing business model.

Create a competitive analysis brief for one or more competitors or a feature area.

Plans and scripts customer discovery interviews using The Mom Test. Use when a user wants to plan a customer interview, write discovery interview questions, build an interview script, generate Mom Test questions, or talk to users about a problem. Produces a non-leading question script grounded in past behavior, plus a note-taking template.

Use when analyzing experiment results, interpreting CSVs from data/raw/ or data/processed/, writing results sections, calculating any of the 12 thesis metrics, running statistical tests, or identifying the crossover point.

Turns a large volume of customer feedback (support tickets, app-store reviews, NPS verbatims, survey free-text) into ranked themes with sentiment, volume, and trend-over-time. Use when you have hundreds or thousands of comments and need to know "what are people complaining about," "what changed since last quarter," or "which theme should we fix first." The quant/at-scale complement to interview-depth synthesis.

Capture the company's house style (voice, formatting, terminology, templates, branding) so every other skill generates in your format. Use when the user says "match our format", "use our style", "here's an example doc — write like this", "set up our house style", "make outputs look like ours", or pastes a sample doc / brand guide to learn from.

Drafts the communications a PM owns when something is breaking or just broke — an internal escalation note, customer-facing status updates on the right cadence, a resolution/all-clear, and a short blameless postmortem for leadership. Use when you say "we have an incident", "the launch broke", "draft a customer notice", "status page update", "outage comms", "customers are hitting errors", "we're down", or "write the postmortem". Reactive comms only — not pre-mortem (failure before launch), release-notes (what shipped), or stakeholder-update (routine status).

Prioritize a set of CX initiatives, ideas, or opportunity areas into an actionable strategic roadmap. Scores each initiative across strategic alignment, customer impact, a 4-bucket feasibility filter (Technical / Operational / Channel / Legal Readiness), and time-to-value. Outputs a three-phase roadmap: Phase 01 Value Realization (quick wins using existing capability), Phase 02 Strategic Enablers (foundational investments for scale), Phase 03 Growth Accelerators (high-impact transformation plays). Change Management is treated as a through-line across all phases. Use when you have more initiatives than capacity and need a defensible, stakeholder-aligned portfolio decision. Trigger when a user asks "where should we start?", "which of these should we do?", "how do we sequence this?", or shares a list of ideas from a workshop or synthesis. Makes explicit trade-offs — refuses to mark everything as high priority.

Use when a PM or knowledge worker wants to think out loud with a sharp sparring partner. Triggers on "brainstorm with me", "be my thinking partner", "stress-test this idea", "explore this problem space", "help me think through", or "challenge my assumptions". Diverges to generate breadth, then pressure-tests and helps converge on a next experiment.

Synthesize user research from interviews, surveys, and feedback into structured insights.

Sorts a pile of inbound feature requests from sales, support, and users into themes, quantifies demand behind each, and recommends build / watch / decline with a templated reply for every requester. Use when you say "I have 40 requests to go through," "what are people actually asking for," "triage this feedback dump," "theme these tickets," or "should we build this or say no?

aStory 블로그 플랫폼을 위한 8개의 AI 저자 페르소나. 아키텍트(격식체 분석적 권위), 개발자(실용적 대화체 구현자), 스토리텔러(서사적 공감적 여정 전달자), 멘토(체계적 양육적 교육자), 분석가(객관적 데이터 기반 연구자), 리뷰어(균형 잡힌 비교 평가자), 큐레이터(간결한 사실적 리포터), 칼럼니스트(비판적 도발적 논평가)를 포함한다. 저자 음성을 선택하거나 페르소나 기반 콘텐츠를 구현할 때 사용한다.