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/emberglow-dev/emberglow-keyboard/claude-mdgit clone --depth 1 https://github.com/emberglow-dev/emberglow-keyboardWrote 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/emberglow-dev/emberglow-keyboard/claude-md)<a href="https://agentmods.dev/instructions/emberglow-dev/emberglow-keyboard/claude-md"><img src="https://agentmods.dev/badge/instructions/emberglow-dev/emberglow-keyboard/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 | $0.01647 | $0.01647 |
| Opus 5 | $0.00823 | $0.00823 |
| Sonnet 5 | $0.00329 | $0.00329 |
| Haiku 4.5 | $0.00165 | $0.00165 |
Grade A, and why
emberglow-keyboard 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — working in the Emberglow repo
Guidance for Claude (and Claude Code) working in this codebase. Read this first.
What this is
Emberglow lights a Keychron Q10 based on Claude activity, over the VIA protocol (QMK raw HID). Two front ends share one lighting engine:
- CLI (
emberglow set <state>) — called by Claude Code hooks. - Flask server (
emberglow serve) — receives Anthropic Managed Agents webhooks, verifies their HMAC signature, and drives the same engine.
Hard-won knowledge (read before touching keyboard.py)
The VIA lighting protocol has two incompatible dialects, and the Q10 speaks the older one. Always handshake the version first — never assume the dialect.
- v3 (protocol version ≥ 11):
set = [0x07, channel, value_id, data…], value IDs1–4, RGB-matrix channel3. - v2 (protocol version ≤ 10 — the Q10 reports 10):
set = [0x07, value_id, data…], value IDs0x80–0x83, no channel byte.
Both use command byte 0x07, and the firmware echoes every packet back even
when it ignores it. So sending v3 packets to a v2 board looks perfect — clean
echo, no error — while doing nothing, and reads return 0 on a visibly-lit
board. The telltales that cracked this: reads were [0] on a glowing keyboard
(brightness can't be 0 on a lit board), and the version line said 10 when v3
would be 11+.
keyboard.detect_dialect() reads version command 0x01 (the one packet
identical across all VIA versions), picks v2/v3 framing, and _verify() does one
read to confirm the reply echoes the request header. This runs before every
command via Keyboard.apply(). Do not bypass it with hardcoded framing — that
is the exact bug the next contributor would reintroduce. The original relay.py
POC is v3-only and does not work on this board; it's kept for reference only.
A second trap: a probe that only changes the effect index is nearly
invisible against whatever the board was already doing. emberglow probe forces
full-brightness red on every step so each effect change is unmistakable.
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 · 126 lines · 1,647 tokens per session scan A 149db34231b0
emberglow-keyboard CLAUDE.md is an instructions file published in the GitHub repository emberglow-dev/emberglow-keyboard (10 stars, last pushed 2mo ago), licensed MIT. It adds 1,647 tokens to every session, about $0.0082 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 2026-08-31.
Other instructions, from other repositories
toh-framework CLAUDE.md
Claude Code instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.
awesome-copilot-id AGENTS.md
AGENTS.md instructions for GulajavaMinistudio/awesome-copilot-id, covering communication, explanation and documentation, markdown formatting, user communication style and workflow & methodology.
antigravity-testing-kit GEMINI.md
Instructions for anhtester/antigravity-testing-kit, covering gemini ai - global automation agent rules, git pull restriction rule, browser rules (mandatory), 🖥️ viewport & mode and 🔄 thứ tự debug bắt buộc (playwright mcp).
sdd-pilot sddp-bootstrap.instructions.md
Instructions for attilaszasz/sdd-pilot: These are project-level bootstrap documents. Edit each via its owning skill.
codevira CLAUDE.md
Instructions for sachinshelke/codevira, covering codevira — persistent project memory, when to call which codevira tool, at the start of every session, before modifying any file and before adopting a pattern, library, or naming convention.
cc-use-exp general.instructions.md
Instructions for doccker/cc-use-exp, a project described as: ⛔ No longer maintained → dev-agent-kit/recipes 让 Claude Code、Antigravity、Gemini CLI、Codex、Cursor 开箱即用的分层配置模板，总结十多年的日常开发经验.