Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/hook-365/adsb-3d/claude-mdgit clone --depth 1 https://github.com/hook-365/adsb-3dWrote 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/hook-365/adsb-3d/claude-md)<a href="https://agentmods.dev/instructions/hook-365/adsb-3d/claude-md"><img src="https://agentmods.dev/badge/instructions/hook-365/adsb-3d/claude-md.svg" alt="Measured on agentmods" 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 | $0.02735 | $0.02735 |
| Opus 5 | $0.01367 | $0.01367 |
| Sonnet 5 | $0.00547 | $0.00547 |
| Haiku 4.5 | $0.00274 | $0.00274 |
Grade A, and why
adsb-3d CLAUDE.md scanned grade A with 0 findings 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 4d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Orientation for an AI assistant working in this repo. Project-specific only —
no deployment/homelab details (those live in env vars and .env, never here).
What this is
ADS-B 3D — real-time 3D visualization of ADS-B aircraft traffic. A single nginx Docker image serves a Vite / TypeScript / Three.js frontend and reverse- proxies a user-supplied ADS-B feeder plus two optional FastAPI services. The frontend uses no framework — DOM and Three.js directly.
Three feature tiers, all optional and off by default: historical playback
(needs track-service + TimescaleDB), ACARS messages (needs acars-service),
and a VHF voice scanner (needs an external voice-services stack).
Repo layout
frontend/— Vite + TS + Three.js app (~37 modules undersrc/). Built to static assets, served by nginx.track-service/— FastAPI + asyncpg. Live WebSocket diff stream + a historical-data collector writing to TimescaleDB.acars-service/— FastAPI bridge to an external acarshub TCP JSON feed.nginx/— reverse-proxy + static-host config templates.Dockerfile+entrypoint.sh— single image. The entrypoint rendersconfig.js(frontend runtime config) and the nginx config from environment variables at container start.docs/—VOICE.md,REVERSE-PROXY.md,PRODUCTION-CHECKLIST.md.
Frontend architecture
The defining pattern is inverted dataflow + a reconciler:
aircraft/store.ts(AircraftStore) is the single source of truth for aircraft state.aircraft/reconciler.tsruns once per frame: it diffs the store against the Three.js scene and creates / updates / removes a per-aircraftGroup(cone, ground icon, altitude line, trail, CSS label, rings). The reconciler owns the aircraft scene graph — do not add or remove aircraft objects elsewhere.
State is shared via subscribe-pattern singletons: a module-level singleton
exposing getX() / updateX() / subscribeX() backed by a listener Set.
No reactivity library. Examples: core/settings.ts, core/theme.ts,
core/time-context.ts, core/filter.ts, feed/feeds.ts,
feed/voice-calls.ts, aircraft/acars-store.ts.
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.
- 4d ago First seen · 205 lines · 2,735 tokens per session scan A f1e737902f72
adsb-3d CLAUDE.md is an instructions file published in the GitHub repository hook-365/adsb-3d (46 stars, last pushed 17d ago), licensed MIT. It adds 2,735 tokens to every session, about $0.0137 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
webxr-first-steps AGENTS.md
Instructions for meta-quest/webxr-first-steps, covering agent instructions — webxr first steps, stack and key facts, build and run, what the sample demonstrates and notes for agents.
webxr-first-steps meta-quest.instructions.md
Instructions for meta-quest/webxr-first-steps: For this repository, prefer Meta Quest Agentic Tools, the Meta Horizon VS Code/Cursor extension, and the hzdb MCP server.
cvr-sdk-unity CLAUDE.md
Claude Code instructions for CognitiveVR/cvr-sdk-unity, covering cognitive3d unity sdk technical reference, critical rules, operating rules, trust hierarchy and pre-implementation discovery.
webxr-first-steps copilot-instructions.md
Instructions for meta-quest/webxr-first-steps: This is a Meta Quest / Horizon OS sample.
webxr-first-steps GEMINI.md
Instructions for meta-quest/webxr-first-steps, a project described as: Your first steps into immersive web development with WebXR.
THREE.Terrain AGENTS.md
Instructions for IceCreamYou/THREE.Terrain, covering agent process, repository areas, randomness, demo changes and validation.