lustre
Skill3 starsChanged 9 months agoMITView on GitHub
Gleam frontend development skill targeting JavaScript. Covers Lustre 5.x, MVU Architecture, Advanced Forms (save-lifecycle state machines/FieldState), UI Patterns, SPA Routing (modem), and HTTP requests (rsvp). Use when building UI or working in client/src/.
In this repository7 other files it gives its agents.
- skills/pg-gleam/CLAUDE.mdCLAUDE.md
- api-gleamSkill
- frontend-testingSkill
- gleam-backendSkill
- gleamSkill
- pg-gleamSkill
- tddSkill
The file
Files
Lustre & Frontend Best Practices
Skill for Gleam frontend development using Lustre. Read the relevant reference before writing code.
Token Efficiency
All agents and commands using this skill MUST follow references/token-efficiency.md (from core). Use Read and Grep tools with targeted reads instead of reading entire files.
Reference routing — by task
| Task | References to read |
| Frontend MVU Forms & UI | advanced-mvu-forms.md + lustre-ui-patterns.md |
| Frontend Web Components | lustre-components.md |
| Frontend HTTP / API calls | lustre-http.md |
| Frontend SPA Routing | lustre-routing.md |
Reference routing — by topic
| Topic | Reference |
| Lustre MVU Core Architecture | lustre-core.md |
| Lustre Effects & Paint cycle | lustre-effects.md |
| Lustre Events (debounce, throttle) | lustre-events.md |
| Lustre Testing (query, simulate) | lustre-testing.md |
| Lustre Browser APIs (plinth) | lustre-browser-apis.md |
| Lustre Advanced (Hydration, FFI) | lustre-advanced.md |
| Common Lustre gotchas | lustre-gotchas.md |
Core Frontend Principles
- Make Impossible States Impossible — Prefer sum types over Bools. Especially for async lifecycles: a form's save is
SubmitIdle | SubmitSaving | SubmitSaved | SubmitFailed(err)(the model owns its data + a siblingsavefield); a page load isRemoteData. Never aBoolflag beside the data. - Parse, Don't Validate — Use
FieldState(v, e)to wrap invalid UI state, rather than dictionaries. - Dumb Effects — Effects should be pure executors of I/O. Never pass the
Modelinto an effect function.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

