Borrowing it
Nothing to install: this file belongs to erisnar/stikart. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/erisnar/stikart/main/CLAUDE.mdgit clone --depth 1 https://github.com/erisnar/stikartWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/erisnar/stikart/claude-md)<a href="https://agentmods.dev/instructions/erisnar/stikart/claude-md"><img src="https://agentmods.dev/badge/instructions/erisnar/stikart/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/erisnar/stikart/claude-md"><img src="https://agentmods.dev/badge/instructions/erisnar/stikart/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03056 | $0.03056 |
| Opus 5.5 | $0.01222 | $0.01222 |
| Sonnet 5.5 | $0.00611 | $0.00611 |
| Haiku 4.5 | $0.00306 | $0.00306 |
Grade A, and why
stikart CLAUDE.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
const fs = require('fs'), { execSync } = require('child_process'); How it starts
The opening of the file, as written. The whole thing — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stikart
Interactive map for Norwegian trail running and ultra race routes. Deployed at stikart.no via GitHub Pages.
Running locally
python3 server.py
# Open http://localhost:8000/index.html
The Python server is needed to avoid CORS errors when fetching local GPX files.
Integration tests
Playwright end-to-end tests in tests/ — the only npm usage in the repo (dev-only; the app itself still has no build step).
npm install && npx playwright install chromium # one-time setup
npm test # run all tests (starts server.py automatically)
npm run test:ui # interactive UI mode
npx playwright test --project=mobile # mobile only
- Two projects:
desktop(Desktop Chrome) andmobile(Pixel 7 emulation — touch events + coarse pointer, soisTouchDevice/isMobilecode paths run) tests/fixtures.jsblocks analytics and serves blank map tiles (offline-safe), and has helpers:touchScrub(real CDP touch events for chart finger-scrubbing),expandPanel,openRaceDeepLink- Tests reference the races
nsm-ultra-2025andlommedalen-rundt(TEST_RACE/DESCRIBED_RACEin fixtures) — update fixtures if those entries are ever removed - Covered: race list/search, race selection + map highlight/dim, elevation chart mouse hover and touch scrub (cursor, km marker, freeze/lock), mobile welcome screen, minimized detail card, deep links
Tech stack
- Vanilla JS, HTML, CSS — no build step, no framework, no npm
- Leaflet.js 1.9.4 for the map
- leaflet-polylinedecorator for route arrows
- Kartverket (Norgeskart) and OpenStreetMap as base layers
- GoatCounter for privacy-friendly analytics
File structure
| File | Purpose |
|---|---|
js/config.js |
Constants: WORKER_URL, isTouchDevice, darkColorPool, raceCategories |
js/races.js |
raceRoutes data array — one entry per race |
js/utils.js |
Pure utility functions: slugify, formatDate, haversineKm, fmtTime, fmtPace, etc. |
js/map.js |
Leaflet map init, tile layers, icons, layer state, loadRace/loadRaces, highlightRace, resetRaceStyles, regenerateColors, race picker popup, custom layer control |
js/filters.js |
Filter state and functions: applyFilters, filterByMonth, filterByCategory, filterBySearch, clearSearch |
js/elevation.js |
buildElevationProfile, renderElevationChart, updateElevCursor |
js/cursor.js |
Shared cursor state (dotFrozen, chartTouchMarker), enableDistanceDot/disableDistanceDot, enableChartMouse/enableChartTouch, minimizeDetail/expandDetail |
js/pace-planner.js |
Pace math (gradeFactor, fatiguedArrival, calcCheckpointSplits), renderPacePlanner, openPacePlanner/closePacePlanner/updatePacePlanner, zoomToCheckpoint |
js/race-panel.js |
Race list panel, race detail overlay, selectRace, showRaceDetailOverlay, closeRaceDetail, shareRace, changeRaceColor; extends applyFilters and regenerateColors |
js/submission.js |
Race submit/edit form: showSubmitRaceForm, openEditRaceForm, handleRaceSubmit, parseGPXForStats |
js/main.js |
Entry point: showInfoOverlay, downloadGpx, startup sequence (regenerateColors, URL params, initial loadRace/loadRaces) |
gpx-parser.js |
Parses GPX XML → GeoJSON; Haversine distance + elevation gain; extracts <wpt> waypoints |
index.html |
Shell: map div, filter bar, panels, overlays |
style.css |
All styles |
server.py |
Local dev server |
worker.js |
Cloudflare Worker — handles race submission/edit PRs server-side (holds GitHub PAT) |
wrangler.toml |
Cloudflare Worker config |
race-calendar/ |
GPX files, one folder per race |
.github/workflows/validate-race.yml |
CI: validates GPX and auto-merges valid race PRs |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 200 lines · 3,056 tokens per session scan A 200dcee2a6c0
stikart CLAUDE.md is an instructions file published in the GitHub repository erisnar/stikart (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 3,056 tokens to every session, about $0.0122 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
stikart CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| stikart CLAUDE.mderisnar/stikart | Instructions file | 2 | 2mo ago | Apache-2.0 | A | 3,056 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 144k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,685 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 150k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,086 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 282k | 3d ago | MIT | A | 4,335 | Codex, OpenCode |