agentsclimarketplace

Lamina business context

Skill aryaniyaps/lamina/skills/lamina-business-context

Headless Product Design skill for AI coding agents | Design how it works, verify what you ship.

Install
npx -y skills add aryaniyaps/lamina --skill lamina-business-context

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Business context UX guidance. Use when bootstrapping Lamina for a project; answering business questions UX work needs; updating context after a pivot or scope change.

SKILL.md

8.2 KB, as published. Nobody here has run it

Business Context

Establishes and maintains .lamina/business-context.md — the business foundation UX workflows read before ideating, specifying features, or optimizing flows.

Guardrail: UX artifacts only. Do not implement product code or visual styling specs.

Load artifacts.md for file contract, changelog rules, and downstream ownership.


Modes

ModeTriggerBehavior
EstablishNo business-context.md, or user requests fresh bootstrapFrame gate → Evidence (brownfield) → Write → Cast
Update/lamina-init update, or explicit pivot/scope-change languageRead existing → re-run changed sections → merge + changelog

Section → skill mapping

Apply the linked capability skill's frameworks when writing each section.

SectionSkills
Problem statementproblem-framing, feature-discovery
Business goalsstakeholder-alignment, product-behavior
Success metricsquantitative-validation, stakeholder-alignment
Scopestakeholder-alignment, feature-prioritization
Users & marketcompetitive-analysis, user-modeling — prose in business-context; structured cast in personas.json during establish
Product postureplatform-posture, product-behavior
Constraintsresearch-scoping, stakeholder-alignment
Stakeholdersstakeholder-alignment
Risks & unknownsfeature-discovery, research-scoping
Research postureproblem-framing, research-scoping
Triad checkproduct-behavior
Inferred context (brownfield only)scan protocol below

Question bank (establish)

Ask one batch of clarifying questions for empty sections — not a multi-step wizard.

Before writing .lamina/business-context.md or .lamina/personas.json, require enough non-placeholder input for Problem statement, Scope, Users & market, Product posture, and Constraints. If any of those core sections are empty or too vague to support downstream UX work, use the clarify output contract and STOP. Only carry unanswered items into Open questions when the user explicitly refuses, skips, or asks to proceed without answering.

Problem statement

  • What user or business problem are we solving?
  • For whom? Why now?
  • Are we building the right thing or validating something we already decided?

Business goals

  • What does organizational success look like in 6–12 months?
  • What would make stakeholders say this project succeeded?

Success metrics

  • Which measurable outcomes matter most (conversion, retention, support volume, task success)?
  • What is the baseline today, if known?

Scope

  • What is explicitly in and out?
  • What would scope creep look like for this project?

Users & market

  • Who are we serving? Who are we explicitly not serving?
  • What alternatives or inertia do users face today?

Product posture

  • Platform (web, mobile, desktop, embedded)?
  • Sovereign vs transient vs daemonic role in users' workflow?

Constraints

  • Time, budget, regulatory, technical, or organizational limits?

Stakeholders

  • Who must support this? Known objections or mandates?

Risks & unknowns

  • What assumptions, if wrong, hurt us in six months?
  • What do we not know yet?

Research posture

  • Generative (find direction) vs evaluative (test solutions)?
  • Which decisions must evidence inform before we build?

Triad check

  • Capability / viability / desirability — which pillar is weakest?

Confidence rubric

LevelMeaning
highDirect stakeholder input, recent validated research, or strong doc evidence
mediumReasonable inference from docs or codebase; not yet validated with users
lowAssumption — flag assumption — needs validation; note in Open questions

Do not present low-confidence assumptions as high confidence.


Staleness rules (update mode)

When a section changes, flag downstream artifacts that may need refresh:

Changed sectionMay stale
Users & marketpersonas.json — re-run /lamina-design discovery and cast update
Scopeprior run.json flows outside new scope; design runs in flight
Business goals, success metricsaudit prioritization; design workflow metrics sections
Product posture, constraintsdesign workflow IA and interaction sections
Problem statement (pivot)personas, prior runs, requirements

Never silently overwrite personas.json or decisions.md. Offer explicit refresh or append.


Brownfield scan protocol

Use when shipped UI or product docs exist. Informs business answers, Inferred context section, and provisional persona cast.

Read order

  1. README, docs/, PRDs, pitch decks, marketing copy
  2. Package manifest / stack signals (framework, app structure)
  3. User-facing routes, pages, layouts, error copy, onboarding — skip tests, build config, unrelated backend unless it affects user behavior

Size heuristics (internal — do not persist)

  • Small: single app, few surfaces → inline read
  • Medium: multiple feature areas, research folders → bounded inline read
  • Large: monorepo or corpus crowds context → bounded repo scan per field-research; summarize in evidence.md

Evidence tagging

Every inferred claim cites @path or states insufficient detail — cannot verify.

Greenfield

Skip scan. Rely on user input and any attached PRD/README.


Artifact template

The section names in the Section → skill mapping are the canonical artifact contract. Write each as its own level-two heading exactly: Problem statement, Business goals, Success metrics, Scope, Users & market, Product posture, Constraints, Stakeholders, Risks & unknowns, Research posture, and Triad check. Additional sections may follow, but must not replace, rename, or combine these headings. Before returning success, validate every canonical section has a non-placeholder **Answer:** line.

---
lamina:
  maturity: brownfield    # greenfield | brownfield
  platform: [web]
  last_updated: 2026-07-06
---

# Business context

## Problem statement
**Answer:** …
**Confidence:** medium
**Evidence:** user input
**Skill:** lamina-problem-framing

## Business goals
…

## Changelog
### YYYY-MM-DD — short label
- Changed: …
- Trigger: …
- Stale: …

Recommended next command (output only)

SignalSuggest
Problem unclear, early exploration/lamina-design
Specific capability to specify/lamina-design
Shipped UI, known pain, clear goals/lamina-verify
Business context incompletefinish open questions before other commands

Do not persist recommendation in any file.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.