
maplibre-source-wiring
Getting a MapLibre GL JS source to actually render — TileJSON url versus hand-wired tiles templates, matching source-layer names to the tile schema, layer order and inserting below labels, promoteId and the feature ids that feature state needs, and the CORS and glyph failures behind a blank map. Use when a source is configured but nothing is drawing, or when setFeatureState does nothing.
- Installs
- 262
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/maplibre/maplibre-agent-skills --skill maplibre-source-wiring- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads maplibre-source-wiring automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
MapLibre Source Wiring
You have a tile URL or a data file and a style, and the map is blank, drawing at the wrong scale, or drawing in the wrong order. This skill covers connecting a source to a style correctly, and the failure modes that look identical from the outside.
When to Use This Skill
The source is already configured and the map is wrong in a way that produces no error. These are symptoms as you would observe them, before you know the cause:
- The map is blank, or a layer you added draws nothing, and the console is clean
- A custom style renders nothing against a tile source that works with the provider's own style
- The basemap is blurry, or sits at a different scale than everything drawn on top of it
- Data you added has covered the street names
- Icons are missing while the rest of the map draws fine
- Hover or click highlighting does nothing at all, and no error is raised
- You have a tile endpoint and are unsure which source type it needs, or whether to point
urlat its TileJSON instead of hand-writingtiles
Not this skill. Text in the wrong font, missing scripts, or self-hosting glyph PBFs — maplibre-fonts-glyphs. Deciding the colors, type, and drawing order of a style you are designing — maplibre-cartography. Deciding whether a dataset belongs in GeoJSON or vector tiles at all — maplibre-tile-sources.
Referencing tiles: url vs tiles
Tiles are addressed by zoom (Z), column (X), and row (Y) — a universal scheme across raster and vector tile sources (see the OpenStreetMap wiki for more information). In a MapLibre source, you reference tiles either directly via a tiles URL template or via a url pointing to a TileJSON endpoint.
When a TileJSON endpoint is available, prefer url. MapLibre fetches the document and reads the tile URL template, zoom range, bounds, attribution, and (for vector tiles) the available source-layers automatically. Tile servers like Martin and tileserver-gl generate TileJSON endpoints for every tileset they serve, as do many hosted providers.
When no TileJSON endpoint exists — for example, a raw raster tile service that gives you a URL template directly — use the tiles array and specify any metadata (minzoom, maxzoom, attribution) in the source definition yourself.
url to a TileJSON endpoint:
{
"type": "vector",
"url": "https://example.com/tiles.json"
}
tiles array:
{
"type": "vector",
"tiles": ["https://example.com/tiles/{z}/{x}/{y}.pbf"],
"minzoom": 0,
"maxzoom": 14
}
{z}, {x} and {y} are tokens MapLibre replaces with each tile's zoom, column and row before it makes a request. A template pasted into a browser with the braces still in it asks for no real tile, so its status code says nothing about the server. To check the server by hand, request a real tile inside the source's zoom range, such as /0/0/0.pbf.
The cost of hand-wiring tiles is that MapLibre has no zoom range unless you supply one, and will assume maxzoom: 22 — requesting zoom levels the tileset doesn't contain, which come back empty. This is why url is the default advice wherever a TileJSON endpoint exists. (For PMTiles specifically the same rule applies and the consequence is sharper — see maplibre-pmtiles-patterns.)
Nothing renders: source-layer and the tile schema
The most common cause of a custom style rendering nothing against a working tile source is a source-layer mismatch.
A vector tile source contains named layers — transportation, water, landuse and so on — and every style layer that draws from it must name one exactly:
{
"id": "roads",
"type": "line",
"source": "basemap",
"source-layer": "transportation"
}
source-layer is required for vector sources and must match the tile schema exactly. A typo or a name from a different schema produces no error and no output — the layer simply draws nothing.
Find the real names in the TileJSON. For vector sources, the TileJSON vector_layers field lists each available source-layer, its attribute fields, and its zoom range. This is the authoritative reference when it is present — TileJSON 3.0 requires it for vector tilesets, but 2.x documents predate it and providers are inconsistent about supplying it. Its absence does not mean there is no schema: identify the schema by name (below) and work from its published layer list. If you use a provider's pre-built style URL, the schema is already matched for you; the mismatch only appears when you write your own layers.
Pre-defined tile schemas
When building a custom style you need to know the tile schema — the source-layer names and their properties. Common schemas:
- OpenMapTiles — the most widely adopted schema, based on OpenStreetMap data. Rich and detailed, with source-layers like
transportation,water,landuse,poi. The largest ecosystem of community styles targets this schema. - Shortbread — an open standard designed to be minimal and interoperable, not tied to any single vendor. Simpler structure than OpenMapTiles; a clean foundation if you're building styles from scratch.
- Protomaps — purpose-built for the Protomaps PMTiles basemap ecosystem. Flat, simple structure with source-layers like
earth,water,roads,places; optimized for serverless delivery. Published layer list: docs.protomaps.com/basemaps/layers.
Each of these publishes its layer list as documentation (see References). None carries a machine-readable version marker, so checking a schema means re-reading the page; nothing can poll it for changes.
When generating tiles with Planetiler or tippecanoe, the output embeds TileJSON metadata in the MBTiles or PMTiles file. Tile servers like Martin read this metadata and expose it as a TileJSON endpoint automatically.
Layer order: drawing below labels
Layers are drawn bottom-to-top in the order they appear in the style. map.addLayer() with no second argument appends the layer above everything, including street names — which is why a data overlay added this way hides the basemap's labels. The fix is to pass the ID of the first symbol layer as addLayer's second argument, found programmatically rather than hardcoded (road-label and friends are schema-specific and break the moment you change basemaps). A raster layer added after vector layers obscures them for the same reason.
For the injection pattern in full, and the canonical layer order for a style built from scratch, see maplibre-cartography.
Missing labels and icons: glyphs and sprite
Two style-root properties supply the assets that symbol layers draw with:
glyphs— URL template for font stacks:"glyphs": "https://example.com/fonts/{fontstack}/{range}.pbf"sprite— base URL for the sprite sheet and metadata, serving both.jsonand.png:"sprite": "https://example.com/sprites/basic"
Pre-built style URLs from hosted providers include their own. When building a custom style or self-hosting, you must supply them.
The two fail differently when absent, and the difference is what makes a symptom readable backwards to a cause. Since GL JS 5.11.0 an absent glyphs does not remove text: MapLibre renders it in a local system font instead, so the symptom is text in the wrong face, not missing text. Reach for glyphs when the labels are there but wrong, not when they are absent entirely. An absent sprite logs a console warning for each missing icon (Image "<id>" could not be loaded) and renders none; map.hasImage(id) returns false for each. A sprite URL that fails to load also logs an error.
For the fallback's mechanism and its limits (it is GL JS only; Native still needs served PBFs), font stacks, self-hosting versus generating, and script coverage, see maplibre-fonts-glyphs.
Raster sources: tileSize and naming traps
Two raster-specific pitfalls that wire without error but render wrong or against the wrong assumption:
tileSize defaults to 512; classic OSM tiles are 256. tile.openstreetmap.org and other classic slippy-map tile servers serve 256px tiles, but MapLibre's raster tileSize defaults to 512px. Omitting tileSize: 256 for a 256px source doesn't error — it renders at the wrong effective zoom level, with imagery and labels appearing a zoom level off, with no error or warning:
{
"type": "raster",
"tiles": ["https://tile.openstreetmap.org/{z}/{x}/{y}.png"],
"tileSize": 256
}
A raster endpoint's name is not its type. An endpoint named "hillshade," "terrain," or similar is not necessarily a raster-dem source. raster-dem requires the tiles to encode elevation as pixel-packed RGB (Mapbox or Terrarium encoding) in PNG or WebP — TileJSON's encoding field, when present, names which. A tile format that can't hold that encoding (JPEG, for instance) with no encoding field is ordinary raster imagery whatever the endpoint is named — a "hillshade" endpoint is often a pre-rendered hillshade image, not raw elevation data for MapLibre to shade client-side. Check the tile format and any encoding field before treating a name as evidence of raster-dem.
Feature state: the source needs an id, and promoteId is usually how it gets one
setFeatureState and the feature-state expression key on a feature's id — a top-level member of the feature object, not a value inside properties. A source whose features have no id gives setFeatureState nothing to attach to, so it fails silently: no console error, the paint expression never fires, and getFeatureState returns empty.
Set promoteId to the property you already key on. It tells MapLibre to use that property as the feature id, and it is the right answer whenever the identifier lives in properties:
map.addSource('parcels', {
type: 'geojson',
data: parcels,
promoteId: 'parcel_id'
});
map.setFeatureState({ source: 'parcels', id: 'APN-1234' }, { hover: true });
For a vector source, promoteId is either a property name applied across all source-layers, or an object of the form {<sourceLayer>: <propertyName>}.[9]
Do not reach for generateId or a hand-written id when a business key exists. Both look like fixes and both fail this case:
generateId: trueassigns ids by index in thefeaturesarray, overwriting any existing values.[9] The ids are not stable across asetDatathat reorders, filters, or appends, so state silently attaches to the wrong feature after an update.- Writing a top-level
idworks only for integers: withoutpromoteId, a feature's id "must be an integer or a string that can be cast to an integer."[9] A key likeAPN-1234is accepted and then never matches. WithpromoteIdthe value may be any primitive.
CORS
If your tiles, glyphs, or sprites are on a different origin, the server must send CORS headers (Access-Control-Allow-Origin). Otherwise the browser blocks the requests and the map is blank or missing labels.
Hosted providers handle CORS for you. For self-hosted servers or static storage, configure CORS on the server or CDN. Range-request sources (PMTiles) additionally need Access-Control-Allow-Headers: Range.
Diagnosing a blank map
Work down this list — the symptoms overlap heavily:
| Symptom | Check first |
|---|---|
| Nothing at all, network shows failed requests | CORS headers; the tile URL itself (404s in the network tab) |
| Nothing at all, network shows 200s | source-layer names against the TileJSON vector_layers |
| Tiles appear then vanish as you zoom in | Missing zoom range on a hand-wired tiles source; use url instead |
| Tiles render, text in the wrong font | glyphs missing, or its URL failing to load |
| Tiles render, no text at all | No text-field on any symbol layer; or GL JS older than 5.11 |
| Tiles render, no icons | Missing sprite at the style root |
| Data draws but hides labels | Layer order; insert before the first symbol layer |
| Vector source draws nothing, raster fine | source-layer missing entirely — it is required for vector sources |
| Raster tiles render a zoom level off | Missing tileSize: 256 on a classic 256px source (default is 512) |
setFeatureState does nothing, no error | Features have no id; set promoteId to the identifying property |
Related Skills
- maplibre-tile-sources — Choosing between GeoJSON and tiles for a dataset.
- maplibre-pmtiles-patterns — Registering the
pmtiles://protocol and PMTiles-specific source setup. - maplibre-fonts-glyphs — Font stacks, glyph endpoints, and script coverage.
- maplibre-sprites-icons — The
spriteURL forms, loading multiple sheets, self-hosting sprite assets, and route shields. - maplibre-cartography — The layer-injection pattern in full, and canonical layer order for a custom style.
References
- MapLibre Style Specification — maplibre.org/maplibre-style-spec/
- TileJSON specification — specification repository on GitHub; version 3.0 (Aug 2021) made
vector_layersrequired for vector tilesets - Slippy map tile naming (Z/X/Y scheme) — OpenStreetMap wiki
- OpenMapTiles schema — openmaptiles.org/schema/
- Shortbread tile schema — shortbread-tiles.org
- Protomaps basemap layers — docs.protomaps.com/basemaps/layers
- Martin tile server (TileJSON endpoints) — maplibre.org/martin/
- MapLibre GL JS docs — maplibre.org/maplibre-gl-js/docs/
promoteId,generateId, andfeature-state— Style Specification: sources and expressions
This skill is a snapshot. Where a primary source contradicts it — the References above, MapLibre's current documentation, or what MapLibre does when you run it — that source wins. Follow it, then report the disagreement, citing the source and your MapLibre version: editing your installed copy helps no one else and is overwritten on the next update.
FAQ
What does the maplibre-source-wiring skill do?
Getting a MapLibre GL JS source to actually render — TileJSON url versus hand-wired tiles templates, matching source-layer names to the tile schema, layer order and inserting below labels, promoteId and the feature ids that feature state needs, and the CORS and glyph failures behind a blank map. Use when a source is configured but nothing is drawing, or when setFeatureState does nothing.
How do I install maplibre-source-wiring?
Run npx skills add https://github.com/maplibre/maplibre-agent-skills --skill maplibre-source-wiring in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/maplibre-source-wiring/ (personal) or .claude/skills/maplibre-source-wiring/ (project).
Which AI agents can use maplibre-source-wiring?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is maplibre-source-wiring safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains maplibre-source-wiring?
It is published by maplibre in maplibre/maplibre-agent-skills. It has 262 recorded installs.
More from maplibre/maplibre-agent-skills
maplibre-tile-sources
maplibre/maplibre-agent-skills
Choosing what form your data needs for MapLibre GL JS — GeoJSON read straight from a file versus vector or raster tiles, the feature-count and file-size thresholds where each stops working, and how a basemap and your own data combine as separate sources. Use when deciding which source type a dataset needs, not when hosting or configuring one.
maplibre-pmtiles-patterns
maplibre/maplibre-agent-skills
Serverless vector and raster tiles with PMTiles for MapLibre GL JS — single-file format, HTTP range requests, hosting on S3/R2/GitHub Pages, generating with Planetiler or tippecanoe, and the pmtiles protocol. Use when you need no tile server or want to host tiles from static storage.
maplibre-mapbox-migration
maplibre/maplibre-agent-skills
Migrating from Mapbox GL JS to MapLibre GL JS — package and import changes, removing the access token, choosing tile sources, plugin equivalents, and what you gain or give up. Use when moving an existing Mapbox map to MapLibre.
maplibre-cartography
maplibre/maplibre-agent-skills
Open the skill to read its SKILL.md instructions.
maplibre-fonts-glyphs
maplibre/maplibre-agent-skills
Open the skill to read its SKILL.md instructions.
maplibre-terrain-rendering
maplibre/maplibre-agent-skills
Open the skill to read its SKILL.md instructions.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available