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/ericjypark/codex-island/claude-mdgit clone --depth 1 https://github.com/ericjypark/codex-islandWrote 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/ericjypark/codex-island/claude-md)<a href="https://agentmods.dev/instructions/ericjypark/codex-island/claude-md"><img src="https://agentmods.dev/badge/instructions/ericjypark/codex-island/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 | $0.03186 | $0.03186 |
| Opus 5 | $0.01593 | $0.01593 |
| Sonnet 5 | $0.00637 | $0.00637 |
| Haiku 4.5 | $0.00319 | $0.00319 |
Grade A, and why
codex-island 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 5d 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Project-specific guardrails. Read every section before touching this repo.
Release process — MANDATORY
This app ships via Sparkle auto-update. Get any of this wrong and you brick auto-update for everyone who's already installed.
The loop (3 commands)
echo "X.Y.Z" > VERSION # 1. Bump
git commit -am "chore(release): bump VERSION to X.Y.Z" \
&& git tag vX.Y.Z # 2. Commit + tag
git push origin main vX.Y.Z # 3. Push (fires CI)
The marketing landing site at ericjypark/codex-island-landing has its own
VERSION file (the hero chip + footer read it at build time). Bump it in
that repo too, in the same release sweep, or the public site keeps showing
the prior version even after brew install ships the new one.
That's it. CI does everything else in ~1.5 min:
- Builds the universal DMG
- Signs it with the EdDSA key from the
SPARKLE_ED_PRIVATE_KEYsecret - Generates
appcast.xmllisting the new version - Uploads DMG + appcast as release assets
- Mirrors the cask to
ericjypark/homebrew-tapwith the new version + SHA-256
Watch with gh run watch --exit-status if you want confirmation, or just trust it.
Hard rules — break these and you brick auto-update
-
VERSIONmust be a single-monotonic version like0.0.X, NOT1or100or anything weird.build.shuses$VERSIONas bothCFBundleVersionandCFBundleShortVersionString. Sparkle comparesCFBundleVersionof the running app againstsparkle:versionin the appcast using Apple's component-wise comparator — so"1"parses as[1]and is larger than"0.0.99". Stay in semver. Always increase. -
The Sparkle public key in
build.sh(SU_PUBLIC_KEY="bz1g...") must NEVER be changed casually. Every existing install verifies updates against this exact key. Change it and every prior install rejects every future update silently. The matching private key lives in (a) the maintainer's macOS Keychain under servicehttps://sparkle-project.organd (b) theSPARKLE_ED_PRIVATE_KEYGitHub Actions secret. To rotate, see the migration note indocs/SPARKLE.md(TL;DR: don't).
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.
- 5d ago First seen · 137 lines · 3,186 tokens per session scan A 70fa20edeeb0
codex-island CLAUDE.md is an instructions file published in the GitHub repository ericjypark/codex-island (318 stars, last pushed 16d ago), licensed MIT. It adds 3,186 tokens to every session, about $0.0159 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
OnlySwitch AGENTS.md
Instructions for jacklandrin/OnlySwitch, covering onlyswitch agent guidance, workflow, context routing and project-specific working rules.
NemoNotch CLAUDE.md
Claude Code instructions for GaoZimeng0425/NemoNotch, covering nemonotch — claude.md, project overview, tech stack, project structure and architecture.
codex-island-app AGENTS.md
Instructions for Jarcis-cy/codex-island-app, covering repository guidelines, 沟通要求, 项目结构与模块划分, 构建、测试与开发命令 and 代码风格与命名规范.
NemoNotch AGENTS.md
AGENTS.md instructions for GaoZimeng0425/NemoNotch, a project described as: An interactive floating panel for the MacBook notch area, turning the notch into a multi-purpose information hub.
FluidAudio copilot-instructions.md
Copilot instructions for FluidInference/FluidAudio, covering fluidaudio - agent development guide, build & test commands, architecture, critical rules and code style (swift-format config).
speech-swift AGENTS.md
AGENTS.md instructions for soniqo/speech-swift, covering agent instructions, communication style, workflow, running tests and benchmarks — sequential and memory-aware and git conventions.