Borrowing it
Nothing to install: this file belongs to AlexandrosGounis/rubber-duckie. 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/AlexandrosGounis/rubber-duckie/main/CLAUDE.mdgit clone --depth 1 https://github.com/AlexandrosGounis/rubber-duckieWrote 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/alexandrosgounis/rubber-duckie/claude-md)<a href="https://agentmods.dev/instructions/alexandrosgounis/rubber-duckie/claude-md"><img src="https://agentmods.dev/badge/instructions/alexandrosgounis/rubber-duckie/claude-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/alexandrosgounis/rubber-duckie/claude-md"><img src="https://agentmods.dev/badge/instructions/alexandrosgounis/rubber-duckie/claude-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 | $0.01136 | $0.01136 |
| Opus 5 | $0.00568 | $0.00568 |
| Sonnet 5 | $0.00227 | $0.00227 |
| Haiku 4.5 | $0.00114 | $0.00114 |
Grade C, and why
rubber-duckie CLAUDE.md scanned grade C with 2 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 12d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- `src/main/claude/` — CLI plumbing: `bin.ts` (PATH-safe binary resolution — GUI apps miss `~/.local/bin`; every spawn must use `claudeBin()` + `spawnEnv()`), `check.ts`/`status.ts` (install + `claude auth status --json` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `src/main/claude/` — CLI plumbing: `bin.ts` (PATH-safe binary resolution — GUI apps miss `~/.local/bin`; every spawn must use `claudeBin()` + `spawnEnv()`), `check.ts`/`status.ts` (install + `claude auth status --json` How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project
Rubber Duckie (package name: rubber-duckie) — a macOS Electron app with two faces:
- Duckie spotlight — an always-on-top liquid-glass panel (the visible app). Type a task and Claude drives the real mouse/keyboard via the
mac_inputnative addon (AX snapshots + CGEvents), narrating progress in the HUD. - Task scheduler — persistent prompts that run through the
claudeCLI unattended on a schedule, with per-run NDJSON logs, reports, diffs, and a session chat. Autopilot rules can auto-queue tasks from GitHub/Jira/Trello/Asana.
Everything executes through the Claude Code CLI — there is no direct API integration and no API key.
Commands
yarn dev # Electron app with hot reload
yarn build # Typecheck + electron-vite build
yarn build:mac # macOS distributable
yarn test # vitest run (tests/ — single file: yarn vitest run tests/scheduler.test.ts)
yarn lint # ESLint with cache
yarn typecheck # typecheck:node + typecheck:web
Use yarn, not npm.
Architecture
src/main/index.ts— lean bootstrap: dotenv (.dev.varsthen.env), windows, tray, runtime, IPC, startup gate.src/main/app/— hidden main window +broadcast()(windows.ts), scheduler/automation singletons (runtime.ts), tray + menu (tray.ts).src/main/ipc/— allipcMainhandlers, one module per domain (claude, onboarding, tasks, computer, integrations, logs, chat, misc).src/main/claude/— CLI plumbing:bin.ts(PATH-safe binary resolution — GUI apps miss~/.local/bin; every spawn must useclaudeBin()+spawnEnv()),check.ts/status.ts(install +claude auth status --json),login.ts(in-appclaude auth login),install.ts(in-appcurl | bashinstaller),runner.ts(detached scheduled runs, stream-json logs),enhance.ts,terminal.ts.src/main/onboarding/— standalone liquid-glass onboarding window + macOS permission checks (Accessibility via the native addon; deep-links to System Settings panes).src/main/computer/— the perceive→act loop (loop.tsorchestrator + split modules), glass HUD windows (overlay.tsbarrel overhud/), native addon wrapper (native.ts).src/main/agents/—AgentScheduler(barrelscheduler.tsover scheduler-core/run-lifecycle/session-runs/…), persistence toagents.json.src/main/integrations/,src/main/automations/— provider OAuth/tasks + Autopilot engine.src/preload/—api.ts(the wholewindow.apisurface) +index.tsbridge; types inindex.d.ts.src/renderer/src/— React 19, plain co-located CSS (no Tailwind).main.tsxroutes#onboardingto the onboarding window; everything else rendersApp. The main timeline window currently stays hidden — the duckie panel and onboarding window are the visible surfaces.src/shared/— types + parsers shared by main/renderer (stream-log, run-report, schedule, claude, permissions, computer).native/src/—mac_input.mm(AX capture, CGEvent input, accessibility checks) andglass_sidebar.mm(NSGlassEffectViewpanels). Build output innative/build/is gitignored.
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.
- 12d ago First seen · 59 lines · 1,136 tokens per session scan C 2252fd5b033a
rubber-duckie CLAUDE.md is an instructions file published in the GitHub repository AlexandrosGounis/rubber-duckie (9 stars, last pushed 1mo ago), licensed MIT. It adds 1,136 tokens to every session, about $0.0057 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).