use-map
Skill68 starsChanged 2 days agoMITView on GitHub
Addresses, routes and delivery order — the project's own map service. Load it when the owner says "show it on a map", "how far is it", "in what order should the courier drive", "find the address", or before you reach for Google Maps and a paid key. The thing nobody guesses from inside the guest layer: this exists, it costs nothing per call, and it answers with real road distances — but only for the region whose map was loaded, and loading one is a heavy background job that the owner starts, not you.
- Sends data out
2 other projects use a version of this.
Adopted by
Most recently taken up by
- Fractera/web4-node-starter-kitword for word, first seen Nov 15, 2026
- Fractera/Agent-Engineering-Infrastructureword for word, first seen Nov 8, 2026
Other versions2 projects use a version of this.
Word for word · 2
- Fractera/Agent-Engineering-Infrastructure★ 68100% same
- Fractera/web4-node-starter-kit★ 68100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository91 other files it gives its agents.
- app/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- adopt-suggested-skillSkill
- ai-elementsSkill
- ai-sdkSkill
- answer-elementSkill
- ask-elementSkill
- audit-broken-charactersSkill
- build-product-with-ownerSkill
- content-factorySkill
- core-build-returnSkill
- core-scheduleSkill
- create-elementSkill
- create-multilingual-content-entrySkill
- dedicated-serverSkill
- design-and-settings-of-elementsSkill
- element-permissionSkill
- element-powerSkill
- expand-site-languageSkill
- explain-this-projectSkill
- find-skillsSkill
- fractera-vs-anthropicSkill
- handle-needSkill
- helpful-linksSkill
- help-in-browserSkill
- humanizerSkill
- integratorSkill
- interface-designSkill
- know-the-architectSkill
- manage-app-settingsSkill
- money-and-resourcesSkill
- node-typeSkill
- own-agentsSkill
- persist-env-var-with-rebuildSkill
- place-page-in-menuSkill
- provider-rulesSkill
- release-templateSkill
- researchSkill
- review-pageSkill
- route-to-elementSkill
- self-reviewSkill
- shadcnSkill
- shape-the-needSkill
- skill-creatorSkill
- study-for-elementSkill
- supervise-elementSkill
- translate-pendingSkill
- ultracode-sessionSkill
- use-agentic-ragSkill
- use-ai-generationSkill
- use-aioSkill
- use-app-configSkill
- use-auth-providersSkill
- use-authSkill
- use-browserSkill
- use-channelsSkill
- use-chatSkill
- use-code-shapeSkill
- use-databaseSkill
- use-dataSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 38
All calls are server-side, through dataFetch.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
use-map
Informational, not binding. Know a better way for the case in front of you — do it your way and say so. You are trusted with the creative work on this project.
The door and the credential are in use-data. Unlike vectors and the knowledge base, there is no
helper for maps in this project yet — you call the door directly with dataFetch. Writing that
helper the first time you need it is welcome; writing a second address or a second key is not.
1. Four questions it answers
🔒 The address is DOUBLED, and this is the trap that a cold run caught (2026-08-23). The proxy
strips exactly /service/<name>, and this service carries a /geo prefix of its own — so the path
is /service/geo/geo/<route>. Written singly it reaches the service as /geocode, which does not
exist there, and the answer is an HTML Cannot POST /geocode — not a JSON error you would recognise.
await dataFetch("/service/geo/geo/optimize", { method: "POST", body: JSON.stringify({ coords }) })The neighbours are not like this: the knowledge base answers at its root (/service/rag/query) and
channels too (/service/channels/status). Ask /capabilities for the prefixes, then check the
service's own route once — a 404 in HTML is what a wrong prefix looks like here.
All calls are server-side, through dataFetch.
| Route | Ask it | Answers |
/service/geo/geo/geocode | { q: "street, house, city" } | { lat, lon, name } · 404 when not found |
/service/geo/geo/route | { coords: [{lat,lon}, …] } in YOUR order | { geometry, distanceKm, durationMin } |
/service/geo/geo/matrix | { coords: [...] }, 2 or more | { distances, durations } — N×N arrays, metres and seconds |
/service/geo/geo/optimize | { coords: [...] } | { order, geometry, totalKm, totalMin } |
🔒 THE UNITS DIFFER BETWEEN NEIGHBOURING ROUTES, and nothing warns you. Measured live
2026-08-23, Eiffel Tower → Louvre: route answered distanceKm: 4.29, durationMin: 10.5, while
matrix answered 4303.9 and 632.1 for the same pair — metres and seconds, straight from the
routing engine. Read a matrix cell as kilometres and you are wrong by a factor of a thousand, in a
number that still looks plausible on a delivery screen.
matrix → {"distances":[[0,4303.9],[4039.4,0]], "durations":[[0,632.1],[610.8,0]]} // m, s
route → {"geometry":…, "distanceKm":4.29, "durationMin":10.5} // km, min
optimize → {"order":[0,1,2], "geometry":…, "totalKm":6.79, "totalMin":17.19} // km, min🔒 The matrix is NOT symmetric. 4303.9 there against 4039.4 back — one-way streets are real.
Averaging the two directions, or reading one cell for both, quietly invents a road that does not exist.
(This section exists because a cold acceptance run had to GUESS the units: the skill said only «raw material for your own logic». A guess that lands on the right answer is still a guess.)
🔒 route keeps your order; optimize chooses one. Sending points to route and calling the
result "the optimal delivery route" is the mistake this pair exists to prevent — it is the length of
the order you happened to pass.
2. 🔒 Real roads, not straight lines — and that is the point
Distances come from a routing engine over real road geometry, so a river without a bridge costs what it really costs. A straight-line ("as the crow flies") calculation of your own is not a cheaper approximation of this; it is a different number that will disagree with the driver.
Nothing here is billed per call: the engines run on this server. That is the argument against reaching for a paid map API for distance work — the key, the quota and the invoice are the things you would be adding, not the capability.
3. 🔒 It only knows the region that was loaded
The map data covers one region at a time — the one the owner provisioned. Ask for a route in another country and the answer is an error, not a longer route.
GET /service/geo/geo/provision-status → { state: "idle" | "downloading" | "processing" | … , region, step }state is a normal part of the product, not an exception. While a region is being prepared, the
service answers with errors; a page that shows "the map is being prepared" is honest, a page that
shows a blank map is broken. Only one provisioning runs at a time — a second request gets 409, with
the name of the region already being built.
🔒 Starting a provisioning is the OWNER's action, from the panel: it downloads and processes several gigabytes and takes a long time. Do not trigger it from application code, and do not offer it as a fix inside a page.
4. What the map service is not
- Not a tile server for the picture. These routes return numbers and geometry. Whatever draws the
map in the browser is a separate decision (and a widget,
use-widgets— with its own rule that the content must exist without JavaScript). - Not a geocoder for bulk imports. It is a single service on the owner's machine; a thousand
addresses at once will be slow and unkind. Geocode when a row is written, store
lat/lonin the row, and never geocode the same address twice. - Not authoritative about addresses.
geocodereturns the first match. For a form where a person types their own address, show them what was found and let them confirm — silently accepting the first match is how parcels go to the wrong street.
5. Before you call it done
routeversusoptimizechosen deliberately, and the step says which.- Coordinates stored on the row, not recomputed per render.
- The "region not ready" and "address not found" paths exist in the product, in the user's language.
- Tested with real points of the loaded region — a route between two invented coordinates proves nothing about the map the owner actually has.
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.

