aryaniyaps/lamina/skills/lamina/SKILL.md
lamina
Route ordinary product implementation and explicit Lamina lifecycle requests. Use passively after initialization for changes to product behavior, flows, permissions, states, failures, or UI; skip purely mechanical maintenance.
- Source repository stars
- 111
- Declared platforms
- 0
- Static risk flags
- 0
- Last source update
- 2026-08-04
- Source checked
- 2026-08-04
Decision brief
What it does—and where it fits
Use this as the compact Lamina router. The public catalog has four workflow skills and six capability skills. Load only the entrypoint and topic references needed for the current request.
Not for
- Tasks that require unconfirmed production actions or broad system permissions.
- Environments where the pinned source and install steps cannot be inspected.
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/aryaniyaps/lamina --skill "skills/lamina"Inspect the Agent Skill "lamina" from https://github.com/aryaniyaps/lamina/blob/af269ef3347fb7edeb956f3af9c943641717891c/skills/lamina/SKILL.md at commit af269ef3347fb7edeb956f3af9c943641717891c. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
Workflow routes
Never recommend /lamina-design or /lamina-verify as a next step in normal flow. Execute the required phase implicitly. Ask a question only when an unresolved decision would materially change the product contract.
/lamina-init: use lamina-init.Ordinary feature, behavior fix, flow refactor, or UI work: prepare the graphExplicit /lamina-design: use lamina-design as a - 02
Ordinary implementation contract
1. Run lamina work prepare without --workflow first. Never invent a workflow ref; narrow only a genuinely ambiguous result with an exact ref. 2. If design gaps exist, run one independent graph-resident Persona walk per active Persona, union discoveries, expand the graph, and rep…
Run lamina work prepare without --workflow first. Never invent aIf design gaps exist, run one independent graph-resident Persona walk perRun lamina work map, resolve every obligation and Persona-bound Experience - 03
Focused-question protocol
For a focused question, do not answer from this router alone and do not suggest the implementation lifecycle. You must:
Open the selected capability's sibling SKILL.md.Open every topic reference needed for the answer.Begin the answer with Using : . - 04
Capability routes
For detailed signal-to-topic routing, read the problem router. Old public names are a hard cutover; use the installed legacy-name lookup to locate their new capability and topic. Read the optional glossary only when unfamiliar Lamina or product-design terminology blocks routing;…
For detailed signal-to-topic routing, read the problem router. Old public names are a hard cutover; use the installed legacy-name lookup to locate their new capability and topic. Read the optional glossary only when unf… - 05
Hard rules
Supporting orchestration is internal to this skill under orchestrator/; it is not a public skill.
Ladybug is canonical and graphd is the sole writer.Never expose raw Cypher or accept caller-supplied epistemic or approval state.Never treat legacy run files, graph dumps, generic screenshots, or broad
Permission review
Static risk signals and limitations
No configured static risk pattern was detected
This is not proof of safety. Runtime behavior, indirect dependencies, and hidden external systems are outside the static scan.
Evidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 85/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 111 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- aryaniyaps/lamina
- Skill path
- skills/lamina/SKILL.md
- Commit
- af269ef3347fb7edeb956f3af9c943641717891c
- License
- Apache-2.0
- Collected
- 2026-08-04
- Default branch
- main
View the original SKILL.md
/lamina
Use this as the compact Lamina router. The public catalog has four workflow skills and six capability skills. Load only the entrypoint and topic references needed for the current request.
Before graph work, read
the CLI prerequisite. /lamina-init
is the one explicit onboarding action. After initialization, ordinary product
implementation language is the primary route.
Workflow routes
/lamina-init: use lamina-init.- Ordinary feature, behavior fix, flow refactor, or UI work: prepare the graph
context, complete design gaps with lamina-design,
derive and check the WorkMap, implement, collect evidence, and run
lamina work verify. - Explicit
/lamina-design: use lamina-design as a graph-only advanced override; never edit application source in that phase. - Explicit
/lamina-verify: use lamina-verify as a source-read-only advanced override; never edit application source in that phase. - Focused product question: select one primary capability below, read its
SKILL.md, then read the smallest relevant topic set before answering. Start the response withUsing <capability>: <topic(s)>so routing is visible and deprecated focused names are never implied.
Never recommend /lamina-design or /lamina-verify as a next step in normal
flow. Execute the required phase implicitly. Ask a question only when an
unresolved decision would materially change the product contract.
Classify wording by requested action, not by the presence of a product problem: “help with,” “what should,” “how should,” critique, explanation, or planning without a request to change repository files is a focused question. Answer it through capability references without requiring initialization or entering the implementation workflow. Requests to add, fix, redesign, refactor, or otherwise change product behavior use the ordinary implementation route.
Focused-question protocol
For a focused question, do not answer from this router alone and do not suggest the implementation lifecycle. You must:
- Open the selected capability's sibling
SKILL.md. - Open every topic reference needed for the answer.
- Begin the answer with
Using <capability>: <topic(s)>.
If the sibling or reference cannot be read, report the missing installed path instead of improvising its guidance.
Capability routes
| Need | Skill |
|---|---|
| Evidence scope, actor walks, grounding, synthesis, research communication | lamina-research |
| Problem framing, business context, priorities, requirements, tradeoffs | lamina-product-discovery |
| Flows, IA, navigation, forms, content, accessibility, feedback, trust | lamina-ux |
| Invariants, dependencies, consistency, concurrency, time, side effects | lamina-product-behavior |
| Structure, feedback loops, traps, leverage, evolution | lamina-systems |
| Persona usability, heuristic review, quantitative validation | lamina-evaluation |
For detailed signal-to-topic routing, read the problem router. Old public names are a hard cutover; use the installed legacy-name lookup to locate their new capability and topic. Read the optional glossary only when unfamiliar Lamina or product-design terminology blocks routing; it is not a second topic index.
Ordinary implementation contract
- Run
lamina work preparewithout--workflowfirst. Never invent a workflow ref; narrow only a genuinely ambiguous result with an exact ref. - If design gaps exist, run one independent graph-resident Persona walk per active Persona, union discoveries, expand the graph, and repeat the full round until every discovery array is empty.
- Run
lamina work map, resolve every obligation and Persona-bound Experience Case to valid implementation and test files, then runlamina work check. The checked map is immutable. - Implement the mapped work.
- Reconcile with one-shot
lamina graph observe, run and publish every active Persona Mission, and passlamina work verifywith the unchanged map.
Every current Persona walk covers every operation node, including denied and inapplicable nodes, and explicitly evaluates authorization, inputs, relationship identity and cardinality, required states, declared Scenarios, Invariant probes, transitions, and all required edge-case axes. Graph closure is authoritative; source retrieval only localizes evidence and code.
Every compiled Experience Case needs one passing oracle event with a structured observation and reproducible artifact. UI obligations require distinct functional, visual, responsive, and accessibility evidence. Missing audit capability blocks verification.
Hard rules
- Ladybug is canonical and
graphdis the sole writer. - Never expose raw Cypher or accept caller-supplied epistemic or approval state.
- Never treat legacy run files, graph dumps, generic screenshots, or broad mappings as substitutes for current graph and Mission evidence.
- Use sessions for multi-fact design changes and preserve stable Resource identity.
- Do not cap active Personas.
- Explicit design and verification phases do not edit application source.
Supporting orchestration is internal to this skill under
orchestrator/; it is not a public skill.
Alternatives
Compare before choosing
aryaniyaps/lamina
lamina
Product design skill for developers who build with AI coding agents. Use when you need to know what to build before prompting your agent — domain model, workflows, edge cases, product states, and UX gaps handled upfront. Works alongside Cursor, Claude Code, Codex, Gemini, Pi. Any stack, any UI library. Do NOT use for visual styling, pixel layouts, or writing app source.
HKUDS/Vibe-Trading
strategy-generate
Create, modify, and optimize quantitative trading strategies, then backtest and evaluate them.
alirezarezvani/claude-skills
quality-manager-qms-iso13485
ISO 13485 Quality Management System implementation and maintenance for medical device organizations. Provides QMS design, documentation control, internal auditing, CAPA management, and certification support. Use when working with medical device quality systems, preparing for ISO 13485 audits, managing regulatory compliance documentation, setting up corrective actions, or building audit preparation programs. Useful for quality management, audit preparation, regulatory compliance, medical device d
MoizIbnYousaf/marketing-cli
higgsfield-generate
Use when the user wants to generate an image or video via Higgsfield AI. Covers 30+ models: Soul V2, Seedance 2.0, Kling 3.0, Veo 3.1, GPT Image 2, Nano Banana 2. Also covers Marketing Studio — branded ad video/image with avatars and products. Use whenever: "generate an image", "make a video", "animate this photo", "image-to-video", "img2vid", "edit this image with AI", "produce a clip", "create an ad", "make a UGC video", "marketing video", "brand video", "TV spot", "import product from URL", "