Borrowing it
Nothing to install: this file belongs to 2dubu/liquid-glass. 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/2dubu/liquid-glass/main/AGENTS.mdgit clone --depth 1 https://github.com/2dubu/liquid-glassWrote 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/2dubu/liquid-glass/agents-md)<a href="https://agentmods.dev/instructions/2dubu/liquid-glass/agents-md"><img src="https://agentmods.dev/badge/instructions/2dubu/liquid-glass/agents-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/2dubu/liquid-glass/agents-md"><img src="https://agentmods.dev/badge/instructions/2dubu/liquid-glass/agents-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 | $0.00850 | $0.00850 |
| Opus 5.5 | $0.00340 | $0.00340 |
| Sonnet 5.5 | $0.00170 | $0.00170 |
| Haiku 4.5 | $0.00085 | $0.00085 |
Grade A, and why
liquid-glass 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 14d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository guidance
This repository distributes one Liquid Glass skill for native iOS, shared by Claude Code and Codex. These instructions govern maintenance of this repository; the installed skill starts at plugins/liquid-glass/skills/liquid-glass/SKILL.md.
Content boundaries
- Cover SwiftUI and UIKit. Keep other Apple platforms and web implementations out of scope.
- Keep
skills/liquid-glass/Markdown-only. Short Swift examples belong inside the relevant reference. Keep test apps, scripts, binary dumps, screenshots, and audit reports outside the repository. - Keep
SKILL.mdconcise: task selection and links to focused references. Add detailed API guidance to the appropriate reference instead of copying it into several files. - Keep all resources needed by an installed skill inside the plugin directory, with relative links. Repository instructions and contributor documentation are not installed skill dependencies.
- Apple Docs MCP and the disassembler are optional investigation tools. Do not add bundled MCP servers, hooks, or executable components without an explicit scope change.
Technical accuracy
- Verify version-sensitive claims against current Apple documentation and the intended SDK. Distinguish Xcode/SDK version, deployment target, and running iOS version/build.
- Keep public contracts, compiler checks, implementation observations, and runtime results separate. A successful compile does not prove rendering, accessibility, or performance.
- Use public APIs in application examples. Do not turn private implementation details into production recommendations.
- When binary inspection is needed, record the OS/build and binary UUID in the external investigation notes. Move only actionable, appropriately qualified guidance into the skill.
- Link claims to their primary sources. Avoid blanket claims that every documented behavior has been runtime-tested or remains current indefinitely.
Packaging
- Preserve the plugin and skill names
liquid-glassand the marketplace install IDliquid-glass@liquid-glass. - Both marketplace catalogs must point to
./plugins/liquid-glass:.claude-plugin/marketplace.jsonfor Claude Code and.agents/plugins/marketplace.jsonfor Codex. plugins/liquid-glass/plugin.jsonis the portable manifest. Keep its shared metadata consistent with.claude-plugin/plugin.jsonand.codex-plugin/plugin.jsoninside that plugin directory.- Keep the portable manifest's
extensions.com.openai.interfaceequal to the Codex compatibility manifest'sinterface. Older Codex clients still need the compatibility manifest. - Bump the version in all three plugin manifests when shipping plugin changes. Keep plugin versions out of marketplace entries so they cannot disagree with the manifests.
- Do not add empty component directories or duplicate the skill for each host. Preserve the README banner and keep install/update instructions aligned with the package.
- Keep repository-root and plugin-root
LICENSEandNOTICEcopies identical so installed plugins retain the license and attribution. Keep the license identifier aligned across all three plugin manifests. - Before changing host integration, consult the current OpenAI packaging guide, Claude manifest reference, and Claude marketplace reference.
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.
- 14d ago First seen · 47 lines · 850 tokens per session scan A 02f8a693e4b5
liquid-glass AGENTS.md is an instructions file published in the GitHub repository 2dubu/liquid-glass (22 stars, last pushed 15d ago), licensed Apache-2.0. It adds 850 tokens to every session, about $0.0034 per session on Opus 5.5. 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 2026-10-02.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).