@ggdna/dna-design-layers 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) ggdna
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,35 @@
+# dna_design_layers
+
+DNA layer: the seven layers of product design — guides and `/layers-*`
+skills.
+
+## What it ships
+
+- `doc/guides/layers-guide.md` — the layer system, design as decision
+  making, how to use the skills, where results are recorded
+- `doc/guides/layers/*-guide.md` — one short guide per layer plus the
+  orient audit
+- `.claude/skills/layers-*` — nine skills that load the matching guide
+
+The guides follow
+[layers-skills](https://github.com/jamiemill/layers-skills) by Jamie Mill
+(MIT). [dna_plan](https://github.com/ggdna/dna_plan) tells where the
+results of a design session land in a project management repo.
+
+## Layers
+
+Orthogonal: this layer carries only its own topic and is combined with the
+other layers by [dna_ggdna](https://github.com/ggdna/dna_ggdna). It declares no
+parent, and it must not — the umbrella lists every topic layer, so taking the
+umbrella back would close a cycle, and a parent here would reach every consumer
+of this topic whether that repo asked for it or not.
+
+## Variables
+
+- `dnaCopyrightHolder` — the name in the license header of every file
+
+## Development
+
+The `dna/` folder is hand-authored source and is never generated. The repo
+instantiates its own DNA — run `dart test` after changes; commit first, a file
+the DNA would overwrite must not carry uncommitted work.

package/dna/_dna.json

@@ -0,0 +1,11 @@
+{
+  // This package is a DNA layer: dna/ is hand-authored source and is the
+  // last layer of its own instantiation.
+  "version": 1,
+
+  // No parent DNA. This layer is orthogonal: it carries only its own topic
+  // and is combined with the other layers by dna_ggdna. It stays parent-less
+  // on purpose — the umbrella lists this layer, so a parent here would close
+  // a cycle, and a parent would reach every consumer of this topic.
+  "layers": []
+}

package/dna/_generated.json

@@ -0,0 +1,101 @@
+{
+  "version": 1,
+  "layers": [
+    {
+      "name": "base",
+      "package": "helix",
+      "ecosystem": null,
+      "resolvedVersion": "1.7.1",
+      "via": null,
+      "hash": "0x6883da335b415f9d"
+    },
+    {
+      "name": "self",
+      "package": null,
+      "ecosystem": null,
+      "resolvedVersion": null,
+      "via": null,
+      "hash": "0x299cd5817d7063f2"
+    }
+  ],
+  "claude": {
+    "claudeMdInclude": null,
+    "claudeMdFromDna": false
+  },
+  "baseVersion": "1.7.1",
+  "baseHash": "0x6883da335b415f9d",
+  "instances": [
+    {
+      "path": ".claude/skills/layers-conceptual-model/SKILL.md",
+      "hash": "0xda240030955579b8"
+    },
+    {
+      "path": ".claude/skills/layers-domain/SKILL.md",
+      "hash": "0xda08f4d45ff49767"
+    },
+    {
+      "path": ".claude/skills/layers-interaction-flow/SKILL.md",
+      "hash": "0x796a950f8a179288"
+    },
+    {
+      "path": ".claude/skills/layers-intro/SKILL.md",
+      "hash": "0x06c1d26e36f43472"
+    },
+    {
+      "path": ".claude/skills/layers-observed-behaviour/SKILL.md",
+      "hash": "0x4108cd91e4e773cd"
+    },
+    {
+      "path": ".claude/skills/layers-orient/SKILL.md",
+      "hash": "0xec099416bd393283"
+    },
+    {
+      "path": ".claude/skills/layers-product-strategy/SKILL.md",
+      "hash": "0xe88cb0659fdbafce"
+    },
+    {
+      "path": ".claude/skills/layers-surface/SKILL.md",
+      "hash": "0x3629176c7b400675"
+    },
+    {
+      "path": ".claude/skills/layers-user-needs/SKILL.md",
+      "hash": "0x78edeebd96aba9bc"
+    },
+    {
+      "path": "doc/guides/layers-guide.md",
+      "hash": "0x83e39ffe86369c5e"
+    },
+    {
+      "path": "doc/guides/layers/conceptual-model-guide.md",
+      "hash": "0x3c0656b53f857fb8"
+    },
+    {
+      "path": "doc/guides/layers/domain-guide.md",
+      "hash": "0x106bb263eb05982a"
+    },
+    {
+      "path": "doc/guides/layers/interaction-flow-guide.md",
+      "hash": "0xd5d3fb1b895d633c"
+    },
+    {
+      "path": "doc/guides/layers/observed-behaviour-guide.md",
+      "hash": "0x056a0821aa714763"
+    },
+    {
+      "path": "doc/guides/layers/orient-guide.md",
+      "hash": "0x8cb92c6b4f366acb"
+    },
+    {
+      "path": "doc/guides/layers/product-strategy-guide.md",
+      "hash": "0x299f8885030cfc02"
+    },
+    {
+      "path": "doc/guides/layers/surface-guide.md",
+      "hash": "0x9281d6ca67379aa9"
+    },
+    {
+      "path": "doc/guides/layers/user-needs-guide.md",
+      "hash": "0x90e00fcd4243afda"
+    }
+  ]
+}

package/dna/_vars.json

@@ -0,0 +1,3 @@
+{
+  "dnaCopyrightHolder": "ggdna"
+}

package/dna/doc/guides/layers/conceptual-model-guide.md

@@ -0,0 +1,67 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Conceptual model guide
+
+Layer 5 of 7, solution space. Skill: `/layers-conceptual-model`.
+
+## Know what this layer is
+
+- The objects the product recognises, their relations, states and
+  vocabulary — independent of any interface
+- The most neglected load-bearing layer; give it more attention than feels
+  comfortable
+- Not the users' mental model, not a database schema
+
+## Decide
+
+- Which objects exist and where their boundaries are
+- How objects relate: cardinality and named roles
+- What a user can do with each object
+- Which states and transitions matter
+- One name per concept, one concept per name
+
+## Keep it honest
+
+- Real objects only: instanceable, structured, useful; "CAC" is an
+  instance of Metric
+- Model an attribute that names another object as a relationship
+- No speculative additions; every element traces to a stated need
+- Can-be-an-object is not should-be-first-class; justify persistence by a
+  concrete flow, mark bets as provisional
+- Split generic verbs that hide different consequences: correct an address
+  vs register a move
+- Redirect implementation talk to what the user can do
+
+## Pick a technique
+
+| Situation                      | Technique                             |
+| ------------------------------ | ------------------------------------- |
+| Material to mine               | Noun foraging, OOUX                   |
+| One object unclear             | Object definition                     |
+| Objecthood uncertain           | Sketch the flow first, then formalise |
+| Relations to check             | Relational object map (`erDiagram`)   |
+| Status changes what is allowed | State diagram (`stateDiagram-v2`)     |
+| Verbs inconsistent             | Action inventory across all objects   |
+| Naming contested               | Ubiquitous language list              |
+| Redesign                       | Walk the existing product             |
+
+Probe the temporal decisions: intermediate states, read lag, relationship
+over time, deletion semantics, history.
+
+## Capture
+
+- Settled object definitions, provisional ones marked
+- Diagrams that encode a real relation or lifecycle
+- Vocabulary decisions with rejected alternatives
+- Open questions for engineering
+
+## Continue
+
+When objects and actions are stable, design how users move through them:
+[interaction flow](interaction-flow-guide.md).

package/dna/doc/guides/layers/domain-guide.md

@@ -0,0 +1,56 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Domain guide
+
+Layer 2 of 7, problem space. Skill: `/layers-domain`.
+
+## Know what this layer is
+
+- What exists in the real world before any product: concepts, terms,
+  processes, mental models
+- Observation, not design
+
+## Decide
+
+- Which key concepts exist and how they relate
+- Which words people use, and where they conflict
+- Where the seams are: groups using the same word differently
+- Which events and processes structure the domain over time
+
+## Keep it honest
+
+- Record contradictions, do not resolve them; resolution belongs to the
+  conceptual model
+- Stay in the real world; push back on product or interface talk
+- Mark nouns as object / attribute / instance / unclear, filter little
+- Say when you map team beliefs instead of researched fact
+
+## Pick a technique
+
+| Situation                       | Technique                      |
+| ------------------------------- | ------------------------------ |
+| Relations unclear               | Concept map (`graph TD`)       |
+| Language contested              | Terminology audit              |
+| Groups diverge                  | Bounded-context mapping        |
+| Process-heavy domain            | Domain event storming          |
+| Knowledge lives in people       | Expert interviews, shadowing   |
+| Domain produces forms, invoices | Document and artefact analysis |
+| Established market              | Competitive analysis           |
+
+## Capture
+
+- Concept map, if it earned its place
+- Terminology conflicts and bounded contexts
+- Key events
+- Noun harvest
+
+## Continue
+
+The noun harvest is the raw material of the
+[conceptual model](conceptual-model-guide.md).

package/dna/doc/guides/layers/interaction-flow-guide.md

@@ -0,0 +1,67 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Interaction flow guide
+
+Layer 6 of 7, solution space. Skill: `/layers-interaction-flow`.
+
+## Know what this layer is
+
+- The places a user moves through, the affordances in each, the content
+  shown, the flow between states
+- Above the model (what exists), below the surface (how it looks)
+- Always for one user in one situation doing one job
+
+## Decide
+
+- Which places the flow moves through
+- What the user can do in each place, and where it leads
+- What content each place shows
+- What happens on failure, empty and edge cases
+- Whether the flow is as simple as it can be
+
+## Keep it honest
+
+- Every affordance has a named destination
+- Edges are required steps: validation, server error, disconnect, timeout,
+  concurrent edit, empty, loading, post-action, cancel
+- No broken objects: attributes and actions of an object stay together
+- No isolated objects: every relation is navigable or deliberately not
+- Vocabulary matches the ubiquitous language
+- Name places for users; more than 5–6 places per job is a signal
+
+## Pick a technique
+
+| Situation                | Technique                 |
+| ------------------------ | ------------------------- |
+| Default                  | Breadboarding             |
+| Finding unmade decisions | Walk the flow as the user |
+| Much to plan, many users | User story mapping        |
+| Redesign                 | Task analysis             |
+| Channels and backstage   | Service blueprinting      |
+| Orientation only         | Flow diagram (`graph LR`) |
+
+Breadboard notation:
+
+```text
+Place name
+- affordance → destination place
+[ content shown in this place ]
+```
+
+## Capture
+
+- The breadboard: places, affordances, destinations, content, conditions
+- A flow diagram, if it aids orientation
+- Open decisions and unresolved edge cases
+- Risks that depend on unsettled lower layers
+
+## Continue
+
+Check that the [conceptual model](conceptual-model-guide.md) is stable,
+then move to the [surface](surface-guide.md).

package/dna/doc/guides/layers/observed-behaviour-guide.md

@@ -0,0 +1,52 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Observed behaviour guide
+
+Layer 1 of 7, problem space. Skill: `/layers-observed-behaviour`.
+
+## Know what this layer is
+
+- What users actually do, said and did — not what the team believes
+- The source of everything above; all other layers interpret it
+
+## Decide
+
+- Which 2–3 questions about users need an answer
+- Which evidence exists and how reliable it is
+- How to gather what is missing
+- Which patterns hold with confidence, which stay assumptions
+
+## Keep it honest
+
+- Stay close to raw data: quotes, actions, feelings
+- Mark every finding: observed / inferred / assumed
+- Name research gaps instead of papering over them
+- Treat workarounds as strong signal
+
+## Pick a technique
+
+| Situation                        | Technique                              |
+| -------------------------------- | -------------------------------------- |
+| No research yet                  | Define the learning goal, then a study |
+| Triggers and motivations unclear | JTBD interviews about a real past case |
+| Saying differs from doing        | Contextual inquiry, shadowing          |
+| Existing product                 | Support tickets, analytics, usability  |
+| Material exists                  | Extract observations, group patterns   |
+
+## Capture
+
+- Key raw observations with their source
+- Patterns with supporting evidence
+- Candidate job stories with confidence
+- Named research gaps
+
+## Continue
+
+Refine the candidate job stories in the [user needs](user-needs-guide.md)
+layer; feed the nouns you heard into the [domain](domain-guide.md) layer.

package/dna/doc/guides/layers/orient-guide.md

@@ -0,0 +1,56 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Orient guide
+
+The front door of the layer system. Skill: `/layers-orient`.
+
+## Know what orient is
+
+- A rapid audit across all seven layers, not a deep dive into one
+- Answers one question: which layer needs attention most urgently?
+- Output: a short audit table and one recommendation
+
+## Ask three questions
+
+- Which product or feature is this about?
+- Which design challenge prompted this?
+- How far along is the work: early exploration, active design, fixing an
+  existing product?
+
+## Audit every layer
+
+Ask one or two targeted questions per layer and rate it:
+strong / partial / assumed / weak / not started / n/a.
+
+| Layer              | Ask                                                  |
+| ------------------ | ---------------------------------------------------- |
+| Observed behaviour | Research, analytics, or team belief?                 |
+| Domain             | How well is the space understood before the product? |
+| User needs         | Can you state the job and why it matters?            |
+| Product strategy   | Which need, which business outcome, explicit?        |
+| Conceptual model   | Clear, shared objects and vocabulary?                |
+| Interaction flow   | Key journeys with places, steps, decision points?    |
+| Surface            | Existing design system or visual language?           |
+
+## Find the bottleneck
+
+- The bottleneck is the lowest layer rated weak, assumed or not started
+- Name it, the missing decisions, and the risk for the layers above
+- Flag assumed layers separately: they look solid and hide the risk
+- Acknowledge deadlines that change the calculus; name the trade-off
+
+## Recommend one next step
+
+- Name the layer guide to work next and why
+- Ask whether to run it now or to push back on the picture first
+
+## Capture
+
+- The audit table
+- The bottleneck and the recommendation, one line each

package/dna/doc/guides/layers/product-strategy-guide.md

@@ -0,0 +1,60 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Product strategy guide
+
+Layer 4 of 7, first layer of the solution space. Skill:
+`/layers-product-strategy`.
+
+## Know what this layer is
+
+- Where understanding turns into choices: which needs to serve, which
+  business outcome to target
+- Bets, not certainties
+
+## Decide
+
+- The business outcome this work serves
+- Which opportunities connect to that outcome
+- Which solution bets to place on them
+- How to test the riskiest assumption cheaply
+- Which bets come first, and why
+
+## Keep it honest
+
+- One measurable, bounded outcome per tree
+- Opportunities are first-person needs anchored to a journey moment, not
+  features; apply the flip test
+- Every opportunity connects to the outcome
+- Every bet names its riskiest assumption; more than one bet per opportunity
+- Every experiment is the cheapest test: days, not months
+
+## Pick a technique
+
+| Situation                   | Technique                              |
+| --------------------------- | -------------------------------------- |
+| Default                     | Opportunity solution tree              |
+| Chosen opportunity          | Solution bets plus experiments         |
+| Many candidate features     | Kano analysis, jobs portfolio          |
+| Choosing the outcome metric | HEART, North Star                      |
+| Shared timeline needed      | Now / next / later roadmap             |
+| Positioning, build or buy   | Wardley mapping, bundling / unbundling |
+| Which flow first            | Critical user journeys                 |
+
+## Capture
+
+- The outcome
+- The opportunity tree, grouped by journey moment
+- The top 2–3 bets with their experiments
+- Deferred bets and untested assumptions
+
+## Continue
+
+The chosen bets define what to model next in the
+[conceptual model](conceptual-model-guide.md). Goals of a quarter live in
+`goals` of the PM repo.

package/dna/doc/guides/layers/surface-guide.md

@@ -0,0 +1,63 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Surface guide
+
+Layer 7 of 7, solution space. Skill: `/layers-surface`.
+
+## Know what this layer is
+
+- Where everything below becomes what a person encounters: words, visuals,
+  motion, feedback
+- Surface problems are often symptoms of deeper layers
+
+## Decide
+
+- Whether the surface honours the model's vocabulary and objects
+- Whether every affordance of the breadboard is present, and nothing exists
+  without a model behind it
+- Whether the emotional register fits the users' jobs
+- How users know what happened, is in progress, or failed
+- What is most prominent, and whether it should be
+- Whether everything is accessible to the users who need it
+
+## Keep it honest
+
+- Tell surface fixes from deeper issues; route deeper ones back to their
+  layer
+- Terms match the ubiquitous language
+- No shapeshifters, no masked objects
+- Errors diagnose, explain and offer recovery
+- Prominence reflects importance
+- Accessibility is decided, not defaulted
+
+## Pick a technique
+
+| Situation                   | Technique                                    |
+| --------------------------- | -------------------------------------------- |
+| Terms feel off              | Vocabulary check                             |
+| Same object looks different | Object-consistency check                     |
+| Something missing or extra  | Completeness check vs breadboard             |
+| Tone misaligned             | Emotional-register check                     |
+| Users unsure what happened  | Feedback and error inventory                 |
+| Too much competes           | Hierarchy review                             |
+| Left to defaults            | Accessibility pass                           |
+| Producing surface           | Mockups, style tiles, prototypes, copy first |
+
+## Capture
+
+- Findings, each tagged surface fix or deeper-layer issue
+- Open surface decisions with options and constraints
+- Cross-layer issues to resolve first
+- What already works well
+
+## Continue
+
+Revisit the surface after every significant change to the
+[conceptual model](conceptual-model-guide.md) or
+[interaction flow](interaction-flow-guide.md).

package/dna/doc/guides/layers/user-needs-guide.md

@@ -0,0 +1,55 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# User needs guide
+
+Layer 3 of 7, problem space. Skill: `/layers-user-needs`.
+
+## Know what this layer is
+
+- What users try to achieve, and why — an interpretation of observed
+  behaviour and domain knowledge
+- Output: opportunities — needs, pains, desires
+
+## Decide
+
+- Who exactly the users are, and in which situation
+- Which jobs they do: functional, emotional, social
+- Which needs are grounded, which are assumed
+- Which needs matter most, and why
+
+## Keep it honest
+
+- State the need, not the solution; push past "I want to export CSV"
+- Strip the mechanism: the "when" must not name your current system
+- Make the "when" picturable: an event, a feeling, a threshold
+- Ask for emotional and social jobs explicitly
+- Mark confidence: observed / inferred / assumed
+
+## Pick a technique
+
+| Situation             | Technique                                  |
+| --------------------- | ------------------------------------------ |
+| Default               | Job stories: when, I want, so I can        |
+| Team works role-based | User stories                               |
+| Large user base       | Top tasks analysis                         |
+| Stakeholder alignment | Persona plus scenario                      |
+| Measurability matters | ODI desired outcomes                       |
+| Many needs            | Prioritise: importance × how poorly served |
+
+## Capture
+
+- Prioritised needs with confidence
+- Surfaced but unprioritised needs
+- Gaps and contradictions between user types
+
+## Continue
+
+The needs are the opportunities of the
+[product strategy](product-strategy-guide.md). If most are assumed, go back
+to [observed behaviour](observed-behaviour-guide.md) first.

package/dna/doc/guides/layers-guide.md

@@ -0,0 +1,65 @@
+<!--
+@license
+Copyright (c) dnaCopyrightHolder
+
+Use of this source code is governed by terms that can be
+found in the LICENSE file in the root of this package.
+-->
+
+# Layers guide
+
+## Know the layer system
+
+- "Layers of Product Design" splits design work into seven layers in
+  three zones; these guides follow
+  [layers-skills](https://github.com/jamiemill/layers-skills) by Jamie
+  Mill (MIT)
+- Lower layers are foundations for upper ones; a weak lower layer creates
+  debt in every layer above
+- Enter at any layer, but check the layers below before building upward
+
+Problem space — knowledge gathered from reality:
+
+- 1 [Observed behaviour](layers/observed-behaviour-guide.md) — what users
+  actually do
+- 2 [Domain](layers/domain-guide.md) — what exists before our product
+- 3 [User needs](layers/user-needs-guide.md) — what users try to achieve,
+  and why
+
+Solution space — deliberate decisions about what to build:
+
+- 4 [Product strategy](layers/product-strategy-guide.md) — which needs we
+  serve, for which outcome
+- 5 [Conceptual model](layers/conceptual-model-guide.md) — which objects,
+  relations and words
+- 6 [Interaction flow](layers/interaction-flow-guide.md) — which places,
+  actions and paths
+- 7 [Surface](layers/surface-guide.md) — what users see, read and hear
+
+## Treat design as decision making
+
+- Every layer makes decisions; artefacts only record them
+- Four kinds of progress: make a decision, uncover an unmade one, evaluate
+  a risky one, prioritise
+- Work one layer at a time; do not mix problem space and solution space
+- Flag bad decisions, not only missing ones
+- Capture decisions, not transcripts; keep the record short
+
+## Use the skills
+
+- Load `/layers-intro` at the start of every design session
+- Run `/layers-orient` when unsure which layer needs attention, see the
+  [orient guide](layers/orient-guide.md)
+- Run the skill of the layer you work on: `/layers-observed-behaviour`,
+  `/layers-domain`, `/layers-user-needs`, `/layers-product-strategy`,
+  `/layers-conceptual-model`, `/layers-interaction-flow`, `/layers-surface`
+- Each skill reads its guide; a guide is a library of techniques, not a
+  script: pick one technique for the live decision, work it, capture the
+  residue
+
+## Record the results
+
+- Write the residue of a session into the project management repo, see the
+  [PM repo guide](pm-repo-guide.md#use-the-design-layers)
+- One decision per file in `concepts/decisions`; everything else in the
+  topic, ux or goal file it belongs to

package/dna/dot-claude/skills/layers-conceptual-model/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-conceptual-model
+description: Define the product's objects, relationships, states and vocabulary independently of any interface
+---
+
+# /layers-conceptual-model
+
+Read
+[conceptual-model-guide.md](../../../doc/guides/layers/conceptual-model-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-domain/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-domain
+description: Map a domain's concepts, terminology conflicts and bounded contexts — the raw material of the conceptual model
+---
+
+# /layers-domain
+
+Read
+[domain-guide.md](../../../doc/guides/layers/domain-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-interaction-flow/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-interaction-flow
+description: Map places, affordances, edge cases and failure paths of a flow without committing to visual form
+---
+
+# /layers-interaction-flow
+
+Read
+[interaction-flow-guide.md](../../../doc/guides/layers/interaction-flow-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-intro/SKILL.md

@@ -0,0 +1,9 @@
+---
+name: layers-intro
+description: Framework orientation for Layers of Product Design — load this first; provides the context all other layers skills depend on
+---
+
+# /layers-intro
+
+Read [layers-guide.md](../../../doc/guides/layers-guide.md) and work by it
+for the rest of the session.

package/dna/dot-claude/skills/layers-observed-behaviour/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-observed-behaviour
+description: Plan user research and synthesise it into confidence-rated findings about what users actually do
+---
+
+# /layers-observed-behaviour
+
+Read
+[observed-behaviour-guide.md](../../../doc/guides/layers/observed-behaviour-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-orient/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-orient
+description: Diagnostic audit across all seven layers — identifies the bottleneck layer and recommends where to focus
+---
+
+# /layers-orient
+
+Read
+[orient-guide.md](../../../doc/guides/layers/orient-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-product-strategy/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-product-strategy
+description: Connect user opportunities to business outcomes and solution bets, and test the riskiest assumptions cheaply
+---
+
+# /layers-product-strategy
+
+Read
+[product-strategy-guide.md](../../../doc/guides/layers/product-strategy-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-surface/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-surface
+description: Audit and decide the surface against the layers below — vocabulary, consistency, feedback, hierarchy, accessibility
+---
+
+# /layers-surface
+
+Read
+[surface-guide.md](../../../doc/guides/layers/surface-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/dna/dot-claude/skills/layers-user-needs/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: layers-user-needs
+description: Elicit and prioritise user needs, pains and desires — the opportunities that feed product strategy
+---
+
+# /layers-user-needs
+
+Read
+[user-needs-guide.md](../../../doc/guides/layers/user-needs-guide.md)
+and work by it. Read
+[layers-guide.md](../../../doc/guides/layers-guide.md) first if this is the
+first layer skill in the session.

package/package.json

@@ -0,0 +1,41 @@
+{
+  "name": "@ggdna/dna-design-layers",
+  "version": "0.1.0",
+  "description": "DNA layer: the seven layers of product design - guides and /layers-* skills",
+  "keywords": [
+    "dna",
+    "helix",
+    "conventions",
+    "scaffolding"
+  ],
+  "homepage": "https://github.com/ggdna/dna_design_layers",
+  "bugs": "https://github.com/ggdna/dna_design_layers/issues",
+  "license": "MIT",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ggdna/dna_design_layers.git"
+  },
+  "type": "module",
+  "files": [
+    "dna",
+    "README.md",
+    "LICENSE"
+  ],
+  "engines": {
+    "node": ">=22"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "scripts": {
+    "test:dart": "dart test",
+    "test": "pnpm run test:dart",
+    "prepublishOnly": "pnpm run test",
+    "format": "prettier --write .",
+    "format:check": "prettier --check ."
+  },
+  "devDependencies": {
+    "@types/node": "^26.4.1",
+    "prettier": "^3.9.6"
+  }
+}