headout-pm-os

The Data Analyst specialist for Headout's PM OS. Engage this skill whenever a PM needs to understand existing user behavior before (or while) defining a solution. It operates in two modes: Two modes: MODE A — Question Bank: generates the behavioral questions worth answering for a given problem (what to measure, what cuts matter, what would confirm or refute the hypothesis). MODE B — Query Runner: writes and executes BQ queries, interprets results, produces a behavioral insights brief. Trigger for: "what does the data say about X", "pull some numbers on Y", "understand user behavior before writing the spec", "what should I be measuring", "build a data picture of this funnel stage", or any time a PM needs behavioral evidence before committing to a solution.

The Experiment Designer specialist for Headout's PM OS. Use this skill when a feature or change needs to be validated via an A/B test or controlled experiment before full rollout. It designs the experiment end-to-end: hypothesis, variants, user assignment, sample size, guardrails, measurement window, and the BigQuery/Statsig setup needed to track results. This skill is conditional — not every feature needs a formal experiment design. Use it when the PM says "we want to A/B test this", "how do we measure if this works", "design an experiment for X", "what's our holdout strategy", "help me think through the experiment setup", or when a spec includes an experiment as its validation method. The Experiment Designer works with Headout's experimentation infrastructure: Statsig for assignment and feature flags, BigQuery for outcome measurement, Mixpanel for behavioral signals, and Delphi (#ask-delphi) for ad hoc queries.

The Idea Generator is the brainstorming specialist for Headout's PM OS. Engage after a Problem Frame is complete — or directly when a PM has a clear problem statement and wants to explore the solution space rigorously before committing to a direction. Acts as a high-fidelity thought partner: builds deep context through a multi-round interview (current state, constraints, prior attempts, platform nuances), maps the solution space into 2-4 strategic themes before generating any ideas, then produces specific ideas within each theme. Trigger for: "what are the ways we could solve this", "brainstorm approaches to X", "what should we build for this problem", "explore the solution space", "I have a problem frame and want ideas", "ideate on this", "what directions could we go", or any request to generate solution directions after a problem has been framed. Output: a structured Idea Brief.

The Implementation Planner specialist for Headout's PM OS. Use this skill to convert an approved PRD and design into a detailed, engineer-ready (and coding-agent-ready) implementation plan. It produces a structured task breakdown with clear sequencing, per-task acceptance criteria, API contracts, state transitions, dependencies, and rollout strategy. The output of this skill is designed to be directly usable by Cursor (or any AI coding agent) as the authoritative source of truth for what to build and in what order. Trigger this skill when a PM says "write the implementation plan for this", "break this into engineering tasks", "create the tech breakdown", "what's the build plan", "prepare this for engineering handoff", or when a spec has passed L1 review and is ready for engineering pickup. The Implementation Planner requires a spec (ideally post-L1 review) as input. It can also incorporate a prototype and data insights brief if available.

The L1 Reviewer is the quality gate for all product specs at Headout before they reach Atish. Use this skill to review any PRD, spec, or requirements document for completeness, logical soundness, scenario coverage, metric clarity, and design coherence. Structured critique, not a polish pass. Rejects incomplete specs with specific, actionable failure reasons. A passing spec is ready for Atish. A failing spec returns to the PM with exactly what needs fixing. Trigger for: "review this PRD", "is this spec good enough", "L1 check", "review before I send to Atish", or whenever the Spec Writer finishes a draft. Checks both the LOGIC layer (scenario coverage, metric rigor, AC quality) and the DESIGN layer (design coherence, prototype alignment).