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/alexasplund/vanchor/agents-mdgit clone --depth 1 https://github.com/AlexAsplund/VanchorWrote 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/alexasplund/vanchor/agents-md)<a href="https://agentmods.dev/instructions/alexasplund/vanchor/agents-md"><img src="https://agentmods.dev/badge/instructions/alexasplund/vanchor/agents-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 5.1 | $0.02154 | $0.02154 |
| Opus 5 | $0.01077 | $0.01077 |
| Sonnet 5 | $0.00431 | $0.00431 |
| Haiku 4.5 | $0.00215 | $0.00215 |
Grade A, and why
Vanchor AGENTS.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 6d 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — start here
You are an AI/LLM about to work on Vanchor-NG. Before editing anything:
1. Read the developer guide
docs/llms/README.md is the curated, LLM-oriented
guide to the whole project. It orients you in ~5 minutes and links to per-area
guides:
docs/llms/architecture.md— data flow, loops, invariantsdocs/llms/backend.md— Python: runtime, control modes, nav, configdocs/llms/simulation.md— physics, sensors, boat parametersdocs/llms/frontend.md— the web UI (VA.*, map, PWA)docs/llms/api.md— REST + WebSocket contractdocs/llms/testing-and-workflow.md— running, testing, gotchas
2. Golden rules
- Simulate, don't theorise. Reproduce + measure control/nav changes in the
harness (
tests/harness.py) before claiming they work. - Keep new defaults a no-op so the full suite stays green.
- Run the server from the repo root (the data dir is cwd-relative).
- Front end has no build step;
node --checkis the only JS gate; the service worker is network-first (bump its version +SHELLlist when you add/change shell assets).
3. 🔁 Keep docs/llms/* current — mandatory
Whenever you add, remove, rename, or change behaviour of code, update the
matching docs/llms/* file in the same change, before considering the task
done. A stale guide misleads the next agent. See the "Keeping these docs
current" section in docs/llms/README.md.
4. Writing docs
For the human docs in docs/ (the docs/llms/* rule above still governs the LLM
guide):
- Concise and spacious. A shorter version with the same information is better. Prefer tables over prose lists. One idea per paragraph.
- One home per topic. Don't duplicate across files — link instead. Adding or
removing a doc means updating the index in
docs/README.md. - Describe current reality, not history.
roadmap.mdis what's next; shipped work belongs in theCHANGELOG. When you implement a design/research doc, fold the still-useful bits into the living docs and delete the stale design doc — don't leave a "recommend X" doc once X ships. Never rewrite theCHANGELOG's dated entries (they're point-in-time). - No stale claims, no broken links. Update wording when behaviour changes;
keep internal
.mdlinks valid. docs/api/is generated bymake docs— don't hand-edit it.
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.
- 6d ago First seen · 152 lines · 2,154 tokens per session scan A bae93dc5db8f
Vanchor AGENTS.md is an instructions file published in the GitHub repository AlexAsplund/Vanchor (118 stars, last pushed 11d ago), licensed MIT. It adds 2,154 tokens to every session, about $0.0108 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
cml-mcp AGENTS.md
AGENTS.md instructions for xorrkaz/cml-mcp, covering agents.md — cml mcp server, project overview, compatibility goal, repository layout and tool modules (src/cmlmcp/tools/).
brilliant_sdk AGENTS.md
Instructions for brilliantlabsAR/brilliant_sdk, covering brilliant sdk — agent guide, how an app works (the pattern behind everything), minimal reading paths, verify without hardware and testing.
NeoMind CLAUDE.md
Claude Code instructions for camthink-ai/NeoMind, covering neomind — edge ai platform for iot, development commands, ecosystem repositories, extension package contract (.nep) and device type template contract (json).
cad-cae-copilot copilot-instructions.md
Copilot instructions for armpro24-blip/cad-cae-copilot, covering github copilot — aieng workspace and essentials.
phone-mcp CLAUDE.md
Claude Code instructions for premex-ab/phone-mcp, covering claude.md, project overview, build commands, architecture and module layout.
zmk-config AGENTS.md
AGENTS.md instructions for urob/zmk-config, covering customization guide, ground rules, how the multi-board layout works, adding a new board and where to change what.