gemini-cli-desktop CLAUDE.md

gemini-cli-desktop CLAUDE.md is an instructions file for coding agents from Piebald-AI/gemini-cli-desktop. It costs 8,628 tokens per session, scanned C, original, MIT.

Repository instructions for Gemini CLI Desktop, a Rust and React/TypeScript desktop and web application for interacting with several command-line AI assistants.

In plain words
What is it for?
Use them when changing the desktop or web app, working with Rust, Tauri, React, WebSockets, AI-provider integrations, or agent-action approvals.
Why use it?
They describe the application’s architecture, technology choices, build and test process, security model, and development workflow.

Instructions file

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/piebald-ai/gemini-cli-desktop/claude-md
Clone the repo
git clone --depth 1 https://github.com/Piebald-AI/gemini-cli-desktop

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 gemini-cli-desktop CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/piebald-ai/gemini-cli-desktop/claude-md.svg)](https://agentmods.dev/instructions/piebald-ai/gemini-cli-desktop/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/piebald-ai/gemini-cli-desktop/claude-md"><img src="https://agentmods.dev/badge/instructions/piebald-ai/gemini-cli-desktop/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 8,628 This file is loaded in full into every session.
When invoked 8,628 The same file — it is already loaded in full.
Security scan C 2 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 5.1 $0.08628 $0.08628
Opus 5 $0.04314 $0.04314
Sonnet 5 $0.01726 $0.01726
Haiku 4.5 $0.00863 $0.00863

Measured 6d ago against content hash 30805ae2ef86, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade C, and why

gemini-cli-desktop 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 6d 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.

Cloud metadata endpointhighServer-side request forgery

One request to 169.254.169.254 can return temporary IAM credentials.

3. Cloud metadata endpoint blocking (169.254.169.254, metadata.google.internal)

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **Network operations**: `curl`, `wget`, `nc`, `nmap`
CLAUDE.md · 906 lines

How it starts

The opening of the file, as written. The whole thing — 906 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides comprehensive guidance to Claude Code (claude.ai/code) when working with code in this repository.

Table of Contents

  1. Project Overview
  2. Architecture
  3. Technology Stack
  4. Development Environment
  5. Build System
  6. Testing Framework
  7. Security Model
  8. API Documentation
  9. Deployment
  10. Configuration
  11. Development Workflow

Project Overview

Gemini CLI Desktop is a powerful, cross-platform desktop and web application that provides a modern UI for Gemini CLI, Qwen Code, and LLxprt Code. Built with Rust (Tauri) and React/TypeScript, it enables structured interaction with AI models through the Agent Communication Protocol (ACP).

Key Features

  • Dual deployment modes: Native desktop app and web application
  • Real-time communication: WebSocket-based event system for live updates
  • Tool call confirmation: User approval workflow for AI agent actions
  • Multi-backend support: Gemini CLI, Qwen Code, and LLxprt Code integration with support for 9+ AI providers (Anthropic, OpenAI, OpenRouter, Gemini, Qwen, Groq, Together, xAI, and custom endpoints)
  • Project management: Session-based workspace management with chat history
  • Security-first design: Comprehensive command filtering and permission system
  • Internationalization: Full i18n support with language switching for English, Chinese Simplified, and Traditional Chinese
  • Custom title bar: Enhanced desktop experience with native window controls
  • About dialog: Integrated help and version information
  • Resizable sidebar: Interactive sidebar with drag-to-resize functionality and persistent width settings
  • Cross-platform support: Windows, macOS, and Linux compatibility
  • File viewing support: PDF, Excel, image, and text file viewers with syntax highlighting
  • Advanced search: Full-text search across chat history and projects
  • Settings management: Comprehensive settings dialog with backend configuration

Read the full file on GitHub · 906 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. 6d ago First seen · 906 lines · 8,628 tokens per session scan C 30805ae2ef86

Subscribe to this mod's changes

gemini-cli-desktop CLAUDE.md is an instructions file published in the GitHub repository Piebald-AI/gemini-cli-desktop (497 stars, last pushed 1mo ago), licensed MIT. It adds 8,628 tokens to every session, about $0.0431 per session on Opus 5. A static security scan graded it C with 2 findings (cloud metadata endpoint, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.