familiar AGENTS.md

familiar AGENTS.md is an instructions file for Codex, OpenCode from dominion525/familiar. It costs 951 tokens per session, scanned A, original, MIT.

A set of coding instructions for Familiar, a Claude Code add-on that controls Google Chrome on macOS with AppleScript.

In plain words
What is it for?
Use it when developing or verifying Familiar's browser-control features.
Why use it?
It explains the repository layout, development process, and conventions needed to modify the add-on safely.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/dominion525/familiar/agents-md
Clone the repo
git clone --depth 1 https://github.com/dominion525/familiar

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for familiar AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/dominion525/familiar/agents-md.svg)](https://agentmods.dev/instructions/dominion525/familiar/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/dominion525/familiar/agents-md"><img src="https://agentmods.dev/badge/instructions/dominion525/familiar/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 951 This file is loaded in full into every session.
When invoked 951 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00951 $0.00951
Opus 5 $0.00476 $0.00476
Sonnet 5 $0.00190 $0.00190
Haiku 4.5 $0.00095 $0.00095

Measured 3d ago against content hash 045bc4831865, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

familiar 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 3d 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.

AGENTS.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

familiar — Agent Guide

A guide for AI coding agents working on this repository. For a human-oriented overview, see README.md / README.ja.md.

Project overview

A Claude Code skill / plugin that drives the real Google Chrome on macOS via AppleScript (Apple Events). It uses neither the DevTools Protocol, Playwright, nor any other separate driver — it operates the Chrome the user already runs every day. The goal is to be treated as a regular user even on pages with bot / WAF defenses that block headless / automated browsers.

Repository layout

  • skills/familiar/familiar.applescript — the script itself (operational core)
  • skills/familiar/SKILL.md — the skill definition. This is the English body read by the skill loader (the auto-activation trigger). Keep it as a thin index and push details into the reference files (progressive disclosure).
  • skills/familiar/SKILL.ja.md — Japanese reference translation (not read by the loader)
  • skills/familiar/reference-browser.md / reference-browser.ja.md — detailed spec for the control surface (tab / window, navigation, waits, content / script)
  • skills/familiar/reference-actions.md / reference-actions.ja.md — detailed spec for the retrieval and interaction actions plus selector strategy (CSS / text= / xpath= / label=)
  • .claude-plugin/plugin.json — plugin manifest
  • .claude-plugin/marketplace.json — marketplace definition
  • README.md / README.ja.md — English / Japanese README
  • AGENTS.md / AGENTS.ja.md — English / Japanese agent guide

Development and verification

  • AppleScript does not need compilation. Run it as-is via osascript.
  • For a syntax check, use osacompile -o /dev/null skills/familiar/familiar.applescript (compiles for syntax only, does not execute).
  • Real verification requires a live Chrome and cannot be done in CI.
  • Prerequisite: enable "Allow JavaScript from Apple Events" in Chrome and accept the automation permission prompt on first run.
  • Invocation form: osascript skills/familiar/familiar.applescript ACTION [ARGS...].
  • For the MCP server (mcp/), install the pre-commit hooks once after cloning: npx lefthook install at the repo root (after cd mcp && npm install). The hooks run Biome, tsc, and Vitest in parallel on staged changes under mcp/.
  • Smoke-testing tip: pages served via data: URLs do NOT execute inline <script> blocks or inline event handler attributes (onclick="...") under Chrome's data-URL security restrictions. When verifying actions like click or set_checked against a data URL, assert against the DOM state the action itself writes (e.g. document.getElementById('check-1').checked) rather than against side effects of inline JS that will not run. Serve from a local file (file://) or a local http server when inline script execution is required.

Read the full file on GitHub · 60 lines

Changes

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.

  1. 3d ago First seen · 60 lines · 951 tokens per session scan A 045bc4831865

Subscribe to this mod's changes

familiar AGENTS.md is an instructions file published in the GitHub repository dominion525/familiar (0 stars, last pushed 2mo ago), licensed MIT. It adds 951 tokens to every session, about $0.0048 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.

Related

Other instructions, from other repositories