AI Agent Documentation

Current Idea Summary

Writes a precise 2–3 sentence definition of the underlying product intention — not the feature label. Strips ambiguity so the team agrees on what is actually being considered.

Why We're Considering It

Surfaces the specific business reason the idea surfaced. Leads with evidence — customer quote, data point, or founder observation — and names names. Never asserts without a source.

User Problem Definition

Writes the precise friction the target user experiences today. One focused paragraph: what they're doing now that's broken, slow, or missing — not a feature gap.

Target User Identification

Calls out the specific persona from the Product Brain. Primary segment first, with one sentence on why they're primary. No generic 'users' language.

Likely Product Area

Names the specific part of the product this idea touches — the actual flow, not just a category. Grounds the idea in the real architecture.

Business Outcome Projection

Identifies which metric this idea moves and by how much if estimable. Ties directly to the Product Brain metrics — not invented success criteria.

Potential V1 Shape

Defines the smallest testable version: one hypothesis, one flow, maximum 3 sentences. Constrained and concrete — a hypothesis engineering can actually test.

Open Questions for the Team

Surfaces 3–5 numbered decisions the team must make, framed as specific team decisions rather than open musings. Prioritizes what is blocking over what is merely interesting.

Important Details Not to Lose

Captures verbatim quotes, constraints, or nuances from context items that typically get dropped in translation. If no context exists, names the evidence that would most raise confidence.

Quality Score & Audit

Scores the idea summary 70–98 on specificity of evidence (30 pts), clarity of problem/user (25 pts), actionability of V1 shape (25 pts), and quality of open questions (20 pts). Includes a 2–3 sentence audit with the one thing the team must confirm before moving forward.

Workflow Type Recommendation

Recommends 'standard' or 'light' workflow based on idea scope, risk, and evidence quality. Standard: new capability, uncertain demand, significant scope, or high strategic stakes. Light: incremental improvement, well-defined scope, strong existing evidence, or low-risk change.

Opportunity Statement

Writes a one-paragraph definition of the opportunity — what it is, who it's for, and why it matters now. Frames it as a strategic bet, not a feature request.

Market Sizing

Estimates Total Addressable Market, Serviceable Addressable Market, and Serviceable Obtainable Market with reasoning tied to Product Brain data and external evidence. Shows methodology, not just numbers.

User Pain Evidence

Compiles and cites the strongest evidence of user pain from context items — direct quotes, usage data, support tickets, research. Separates strong evidence from weak signals.

Competitive Landscape

Maps current alternatives, direct competitors, and whitespace. Identifies the differentiation angle and what would make this solution meaningfully better than what already exists.

Strategic Fit Assessment

Evaluates alignment with the company's strategic focus, where-we-will-not-compete boundaries, and roadmap themes from the Product Brain. Flags misalignment explicitly.

V1 Concept & Scope Boundaries

Defines the V1 concept — what's in, what's explicitly out, and what belongs in V1.1. Gives engineering early scope signal and prevents scope creep before design begins.

Validation Assumptions

Lists the 3–5 key assumptions that must be true for this opportunity to be real. Ranks them by risk. Connects each assumption to a validation approach.

Recommended Next Steps

Identifies the highest-leverage next action: run validation, go to design, do more research, or kill the idea. Gives a clear recommendation with reasoning.

Vision Alignment Score

Scores 0–10 how closely this idea aligns with the company's declared product vision. Cites specific vision language and explains how the idea advances or diverges from it.

Strategic Fit Score

Scores 0–10 alignment with current strategic focus and roadmap themes. Flags ideas that are technically valid but strategically off-course.

Customer Pain Score

Scores 0–10 the strength of evidence that real users experience this problem. Distinguishes between cited evidence and inference.

Lead Generation Value Score

Scores 0–10 the idea's potential to attract new customers or expand market reach. Connects to acquisition metrics defined in the Product Brain.

Weekly Engagement Value Score

Scores 0–10 the potential of this idea to drive recurring, habitual use. Distinguishes 'nice to have once' from 'come back every week.'

Differentiation Score

Scores 0–10 how uniquely this idea strengthens the product's competitive position. Evaluates against alternatives from Atlas's competitive landscape.

Brand Fit Score

Scores 0–10 alignment with the company's tone, voice, and brand rules. Flags ideas that would create brand confusion or inconsistency.

Revenue Value Score

Scores 0–10 the potential revenue impact — new revenue, expansion revenue, or retention improvement. Ties to metrics that matter from the Product Brain.

Build Complexity Score (inverted)

Scores 0–10 (10 = low complexity). Assesses implementation difficulty based on architecture notes, integrations required, and scope from the Opportunity Brief.

Scope Risk Score (inverted)

Scores 0–10 (10 = low risk). Evaluates how likely this idea is to expand beyond the approved V1 scope once development begins.

Integration Risk Score (inverted)

Scores 0–10 (10 = low risk). Assesses the risk from external systems, third-party dependencies, or architectural debt that this idea touches.

Composite Score & Go / No-Go Signal

Combines the 11 dimension scores into a 0–100 composite. High-scoring ideas get a green Go signal; low-scoring ideas get a red No-Go. Mid-range scores trigger a 'Go with Conditions' recommendation.

Risk Flags

Surfaces the 2–3 highest-risk dimensions with specific reasoning and what would need to be true for the risk to be resolved before moving forward.

Strategic Pillar Generation

Synthesizes the Product Brain into 3–5 named strategic pillars — each a distinct area of product investment with clear purpose and scope. Pillars become the organizing framework for every idea.

OKR Alignment Mapping

Maps each strategic pillar to the company's declared OKRs. Shows which objectives are well-supported by current pillars and which have no pillar coverage.

Gap Analysis

Identifies strategic gaps: areas where the vision, personas, or metrics demand investment but no pillar or roadmap initiative currently covers it.

Impact Metric Assignment

Assigns 1–3 measurable impact metrics to each strategic pillar. Metrics are tied to the 'metrics that matter' from the Product Brain — not invented.

Portfolio Coverage Heatmap

Shows which pillars have too many ideas, which are under-resourced, and which have strong evidence vs. weak signals. Helps PMs balance the portfolio before committing resources.

Strategy Insights (Nova-powered)

Generates 4–8 severity-sorted strategic insights using insight types: strategy_at_risk, metric_needs_attention, strong_momentum, coverage_gap, execution_drift, validation_needed, and resource_bottleneck. Each insight includes severity (critical/high/medium/low) and a recommended action.

The Vision

Writes the single declarative vision statement — 1–2 sentences that capture what the product ultimately achieves for the world. Specific, memorable, and directionally clear.

Why This Exists

Articulates the founding insight: the specific problem, market moment, or human need that makes this product necessary now. Grounds the vision in a real observation, not a generic mission statement.

Who It's For

Defines the primary beneficiary in specific, human terms — drawing from Product Brain personas. Includes who it's explicitly NOT for to sharpen focus.

What We're Building

Describes the product in terms of experiences and outcomes, not features. Covers the horizon of the full vision, distinguishing today's reality from the intended future state.

The Bet

Names the core strategic bet: the thing the company believes that others don't, the insight about user behavior or market structure that the product is built on.

The North Star

Defines the single metric or outcome that, if achieved, would confirm the product vision is working. Connects to the 'metrics that matter' in the Product Brain.

Before / After Experience Story

Documents the full contrast between what the user does today (broken, slow, frustrating) and what they'll be able to do after this feature ships. Written as a narrative, not a feature list.

JTBD Unpacking

Translates the Jobs To Be Done into concrete experience implications — what the user needs to feel, accomplish, and avoid at each stage of the flow.

Moments That Matter

Identifies 3–5 pivotal moments in the user experience that will determine whether the feature succeeds or fails. For each: what the moment is, what the user is feeling, and what the design must achieve.

Experience Principles

Generates 4–6 named, actionable design principles specific to this feature. Not generic UX guidelines — principles derived from the specific user context, brand rules, and strategic stakes of this idea.

Flow Narrative

Writes the end-to-end flow as a user narrative — what happens, in order, from the trigger to the success moment. Identifies branching points, error states, and recovery paths.

Prototype Direction

Describes what the prototype must validate, which interactions are highest risk, and what questions should be answered by watching a user attempt the experience. Feeds directly into Pixel's prompt generation.

Brand-Aligned Visual Spec

Incorporates the company's exact colors, typography, component style, spacing, and tone from the Product Brain design system. The prototype looks like it belongs to the product — not a generic template.

Screen-by-Screen Component Spec

Breaks the experience strategy into specific screens, each with a list of components, their states, interactions, and content requirements. Organized in build order.

Interaction & State Definitions

Defines every interactive state — loading, empty, error, success, hover, disabled — for each component. Prevents the prototype from feeling half-finished during user testing.

Claude Code Ready Output

Formats the prompt to work directly as a Claude Code build brief, with the right level of specificity for AI-assisted development: outcome-focused, not implementation-prescriptive.

Prototype Version Iteration

When iterating on an approved prototype, generates a targeted update prompt that preserves what was validated and changes only what Echo's feedback synthesis identified for improvement.

Feedback Theme Identification

Groups raw feedback from multiple users into coherent themes. Separates signal from noise — identifies patterns that appear repeatedly vs. one-off preferences.

Accept / Defer / Flag Decisions

For each feedback theme, makes a recommendation: Accept (implement in next iteration), Defer (valid but out of V1 scope), or Flag (contradicts other feedback or experience strategy — needs team decision).

Contradiction Detection

Identifies when two users want opposite things, or when user feedback conflicts with the validated experience strategy. Surfaces these explicitly rather than silently picking one side.

Priority Fixes List

Ranks the accepted feedback items by impact on the moments that matter from Sloane's strategy. The top 3–5 items are the iteration targets for the next prototype version.

Next Iteration Brief

Generates a specific, actionable brief for Pixel — what to change, what to preserve, and which interactions to validate in the next round of testing.

Feedback Quality Assessment

Evaluates the quality of the feedback session itself — were the right users tested, were the right scenarios covered, are there gaps in the feedback that need a second round?

Decision Package Compilation

Assembles everything the decision-making team needs — the validated opportunity, the approved experience strategy, the prototype state, and all prior decisions — into a single readable document.

Risk Inventory

Surfaces every identified risk from across the pipeline — strategic risks from Nova, UX risks from Echo, scope risks from Atlas, and any flags from the decisions log. Ranks by severity.

Go / No-Go Recommendation

Generates a clear recommendation with explicit reasoning: Go (proceed to Dev & Build), No-Go (kill or park the idea), or Go with Conditions (proceed but resolve specific issues first).

Conditions & Blockers List

For conditional Go decisions, generates a specific list of conditions that must be met before engineering begins. Each condition has an owner and a resolution path.

Stakeholder Brief

Writes a 3–5 paragraph executive summary of the idea's journey — from raw input through validation to the current checkpoint state — formatted for leadership or board presentation.

Scope Confirmation

Confirms the final V1, V1.1, V2, and Not-Doing scope boundaries before Handoff begins. Prevents scope drift between design approval and engineering kickoff.

Product Requirements Document (PRD)

Generates a 14-section PRD covering: executive summary, background, user problem, feature requirements with acceptance criteria and edge cases, user flows, screens & components, data requirements, integration requirements, performance requirements, security & privacy, analytics, roadmap context, launch criteria, and open questions for engineering.

Feature-Level Acceptance Criteria

Writes Given/When/Then acceptance criteria for every feature, covering happy path, error states, loading states, empty states, and permission states. Minimum 3 ACs per feature — often 6–10.

Edge Case Documentation

Identifies and documents every edge case per feature: network failures, data state extremes, concurrency issues, input extremes, session edge cases, and unusual-but-valid business logic scenarios.

Analytics & Measurement Spec

Generates a complete analytics spec with measurement hypothesis, primary/secondary/anti-metrics, event naming convention, and a full event specification for every trigger in the feature — with required properties, types, and implementation notes.

Conversion Funnel Definitions

Maps each key user flow as a funnel with expected drop-off rates per step. Includes interpretation guidance: 'if drop-off at step X exceeds Y%, investigate [specific hypothesis].'

Review Cadence & Decision Framework

Generates a Day 3, Week 1, Week 2, Day 30, and Day 90 review cadence with specific measurement targets and decision triggers (accelerate, monitor, investigate, respond) for each checkpoint.

Technical Handoff with User Stories

Generates a full technical handoff including user stories by epic (As a / I want to / So that format, with story points and priority), architecture overview across frontend/backend/data/infrastructure, and API specifications for every new and modified endpoint.

Database Schema & Migration Plan

Documents schema changes as migration intent — what each migration does, why, impact if it fails, migration risk level, rollback plan, and whether a data backfill is required.

Build Sequence & Parallelization Plan

Recommends a phase-by-phase build order, identifies what frontend and backend work can run in parallel, and provides a complexity estimate table per component with effort, risk, and key considerations.

Claude Code Build Brief

Generates a narrative build brief formatted for Claude Code — outcome-focused, not implementation-prescriptive. Describes what to achieve and why, so Claude Code can read the existing codebase and make implementation decisions itself.

QA Plan with Checkbox Test Cases

Generates a complete QA plan with test suites for happy path (minimum 5 cases), error states, edge cases, empty states, mobile (iOS + Android), performance, analytics event verification, accessibility (keyboard nav, screen reader, contrast), and regression. Every test case uses markdown checkboxes for interactive tracking.

Persona-Specific Journey Tests

For each persona defined in the Product Brain, generates an end-to-end QA journey specific to that persona's entry point, goal, and definition of success with the feature.

Build vs. Spec Comparison

Systematically compares the live build against the approved PRD acceptance criteria and QA Plan test cases. Documents which requirements passed, which failed, and which are not yet testable.

Pass / Fail Matrix

Generates a structured pass/fail matrix organized by test suite: happy path, error states, edge cases, empty states, mobile, performance, analytics, accessibility. Visual overview of build completeness.

Bug Prioritization

For every failing test case, creates a bug report with severity (critical/high/medium/low), reproduction steps, expected vs. actual behavior, and a recommended fix priority before launch.

Launch Readiness Assessment

Evaluates the build against the PRD's launch criteria checklist. Identifies which criteria are met, which are blocking launch, and which could be deferred to a hotfix.

Regression Report

Checks the adjacent areas identified in the QA Plan for regressions. Flags any existing functionality that was broken by this build.

Launch Notes

Writes product launch notes for the team — what shipped, what it does, who it's for, and what changed from the pre-launch state. Designed to be forwarded directly.

Positioning One-Pager

Generates a one-page positioning document: the problem, the solution, who benefits, key differentiators, and how to talk about it. Can be used for sales, marketing, or investor communications.

Comms Pack

Writes ready-to-send communications for relevant channels — customer announcement email, internal team announcement, changelog entry, social media post. Each adapted to the right voice for its audience.

Adoption Plan

Builds a 30-day adoption playbook: who to target first, what activation looks like, how to onboard users to the new feature, and what early success signals to watch for.

Measurement Plan

Generates the Day 3, Week 1, Week 2, Day 30 measurement plan from the Analytics Spec — who reviews it, what numbers to look at, and what decisions each checkpoint triggers.

Post-Launch Readout

After 30 days, generates a structured readout: what the metrics show, how they compare to targets, what the feature achieved against the original hypothesis, and what the recommended next move is (accelerate, iterate, or pivot).

Evidence Retrieval

Searches the transcript library for exact mentions of a topic, persona, feature, or pain point. Returns verbatim quotes with source transcript and speaker attribution.

Pattern Recognition

Identifies how often a topic appears across transcripts, whether it clusters around specific personas, and whether frequency has increased or decreased over time.

Sentiment Classification

Classifies mentions as positive, negative, neutral, or mixed. Helps distinguish between features users complain about, features they celebrate, and features they're indifferent to.

Persona Segmentation

Filters research results by the personas defined in the Product Brain. Answers: 'What do enterprise customers say about X versus what SMB customers say?'

Evidence Summary

Synthesizes all retrieved evidence into a 2–3 paragraph research brief that can be attached to any idea as a context item — turning raw transcripts into structured input for Mira, Atlas, and Nova.

Research Gap Detection

Flags when a question can't be answered from the existing transcript library — identifying where new research is needed before validation is possible.

Multi-Format Parsing

Parses ServiceNow demands, Jira exports, ADO dumps, Linear boards, spreadsheets, and strategy documents. Normalizes fields across different source formats into the Product Dev OS data model.

Duplicate Detection

Identifies duplicate or near-duplicate items across sources — common when importing from multiple tools or when the same initiative was captured in Jira and ServiceNow. Merges duplicates and preserves provenance.

Agentic Object Organization (Step 3.5)

After ingest, Harbor autonomously organizes imported objects into coherent groups before the user reviews them. Clusters by theme, strategic pillar, and value similarity — dramatically reducing manual sorting time.

Scoring & Prioritization

Applies a scoring model to each ingested item based on business value indicators, urgency signals, strategic alignment keywords, and effort proxies. Produces an initial ranked list for human review.

Metadata Enrichment

Extracts and attaches metadata from source records: original ID, source system, timestamp, requestor, priority label, epic or theme tag, and any free-text notes. Preserved alongside the normalized record.

Portfolio Modeling

Groups the ingested portfolio by theme, strategic area, or team alignment. Gives PMs a portfolio view immediately after ingest — not a flat list but a structured breakdown ready for decision-making.

Capacity Analysis

Calculates available engineering capacity by role and seniority across a given planning period. Accounts for meetings, reviews, on-call, and planned time-off. Returns realistic net hours per person.

Initiative Fit Assessment

Maps each initiative's estimated effort against available capacity. Shows clearly what can fit in the plan, what can't, and what is borderline — requiring a scope or timeline decision.

Bottleneck Detection

Identifies where demand exceeds supply — specific roles (e.g., 'senior backend is at 140% capacity'), specific teams, or specific time windows. Shows the exact constraint before it becomes a missed deadline.

Date Optimization

Given a set of initiatives and constraints, recommends the optimal sequencing and start dates. Identifies which initiatives can run in parallel and which create dependencies that determine the critical path.

Trade-off Modeling

Models 2–3 'what if' scenarios: what happens if Initiative A is descoped, what changes if one engineer is added, what if X is delayed by 2 weeks. Gives decision-makers concrete data for trade-off conversations.

Capacity Narrative

Summarizes the capacity situation in plain language for leadership — not just a spreadsheet but a 3–5 sentence readout of what the team can realistically deliver and what would need to change to deliver more.