by tanaikech
High-performance Google Drive CLI and Model Context Protocol (MCP) Server for LLM/AI agents. Natively execute Google Apps Script (GAS) under a secure whitelisted runtime sandbox, automate transfers, and manage Drive infrastructure.
# Add to your Claude Code skills
git clone https://github.com/tanaikech/ggsrunLast scanned: 6/1/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-06-01T09:27:21.276Z",
"npmAuditRan": true,
"pipAuditRan": true
}ggsrun is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by tanaikech. High-performance Google Drive CLI and Model Context Protocol (MCP) Server for LLM/AI agents. Natively execute Google Apps Script (GAS) under a secure whitelisted runtime sandbox, automate transfers, and manage Drive infrastructure. It has 171 GitHub stars.
Yes. ggsrun passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/tanaikech/ggsrun" and add it to your Claude Code skills directory (see the Installation section above).
ggsrun is primarily written in Go. It is open-source under tanaikech on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh ggsrun against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.

ggsrun is an enterprise-grade CLI application and Model Context Protocol (MCP) Server designed to relentlessly orchestrate Google Drive I/O operations and statefully execute Google Apps Script (GAS) natively from your local terminal or via autonomous Large Language Model (LLM) agents.
Built on Go 1.26.4+, ggsrun transitions legacy single-threaded processing into a state-of-the-art, channel-based concurrent streaming architecture. It delivers maximum network throughput, native Shared Drives (Omni-Drive) support, advanced MIME resolution, resilient self-healing retries, and a native Security Sandbox to restrict Google Workspace APIs during autonomous executions.
To make onboarding, development, and advanced operations as seamless as possible, our documentation is fully modularized. Navigate directly to your area of interest:
| Guide / Manual | Description | Link |
|---|---|---|
| 🚀 Setup & Onboarding Guide | Step-by-step instructions to configure GCP, generate OAuth2 client secrets, link Google Apps Script, and deploy gateway endpoints. | Setup Guide |
| 📖 Command Reference Manual | In-depth breakdown of all 15+ subcommands (e.g. exe1, exe2, download, upload), recipes, and Mermaid architectural diagrams. |
Command Reference |
| 🛡️ Security Sandbox Guide | Explains the memory-based wrapper injection, whitelist configuration schemas (sandbox_config.json), and security validation scenarios. |
Sandbox Guide |
| 🔄 Stateful Execution Lifecycle | Deep-dive explanation of the backup, sandboxing, upload, execution, and resilient rollback phases of the exe1 command. |
Execution Lifecycle |
| 🤖 MCP Server Guide | Configuration to run ggsrun as an autonomous tool provider in AI environments (like Claude Desktop or Antigravity), schemas, and scenarios. |
MCP Server Guide |
| 💻 Interactive TUI Filer Guide | All keyboard shortcuts, search highlighter rules, clipboard integration, and history for the split-screen Japanese PC-98 Filer Mode (ggsrun fd). |
TUI Filer Guide |
| 🧪 Local Development & Testing | Detailed guide for configuring environment variables (.env), understanding CLI/TUI mock tests, and running automated test suites. |
Development Guide |
| 🔬 Manual Integration Tests Suite | Guided verification commands to manually test authorization, sandbox policies, tool schemas, and local file operations. | Manual Tests Suite |
| 📚 Detailed User Manual | Comprehensive manual detailing all commands, advanced options, custom editor integrations (Sublime Text), and legacy parameters. | Detailed Manual |
| 📜 Version Update History | Chronological record of all updates, bug fixes, features, and refactoring milestones from v1.0.0 to the latest release. | Update History |
--accesstoken / --at) for automated testing and CI/CD pipelines.--sandbox) to guard Workspace resources.ggsrun recover).HTTP_PROXY, HTTPS_PROXY, NO_PROXY) for robust connectivity within Gemini Managed Agents sandboxes and enterprise corporate networks.Legacy single-threaded processing has been completely replaced. ggsrun implements a channel-based worker pool built on golang.org/x/sync/errgroup to maximize network throughput on massive folder hierarchies.
The v5 engine forces supportsAllDrives=true and includeItemsFromAllDrives=true across all Google Drive API permutations, allowing enterprise Shared Drives to be mapped and managed seamlessly.
The extraction logic dynamically categorizes Google Workspace entities. Downloader rules bypass the standard Drive API for GAS files, automatically routing to the Apps Script API to download scripts natively as structured JSON or packaged ZIP archives.
The v5 execution and transfer phase is strictly non-blocking. Google API Rate Limits (HTTP 429) and Server Errors (5xx) trigger an exponential backoff sequence per-worker, ensuring resilient self-healing.
Running ggsrun mcp transforms the application into an autonomous JSON-RPC background server via standard I/O, allowing Large Language Model (LLM) agents to search, transfer, and execute scripts safely.
Requires Go 1.26.4 or higher installed. Compile and install the binary natively:
$ go install github.com/tanaikech/ggsrun@latest
Alternatively, download the pre-compiled binary matching your CPU architecture and operating system from the Official Releases Page:
ggsrun_darwin_amd64 (Intel) or ggsrun_darwin_arm64 (Apple Silicon M1/M2/M3)ggsrun_linux_amd64 (or matching ARM, 32-bit, or MIPS variants)ggsrun_windows_amd64.exe (or 32-bit, ARM variants)Once you have downloaded or compiled ggsrun, verify that the binary is functional inside your terminal:
Run the help display command to verify the binary executes and lists available options:
$ ggsrun --help
Before running any diagnostics or script execution commands, you must configure your Google Cloud credentials and authorize the local application.
Please follow the detailed, step-by-step instructions in the Setup & Onboarding Guide to:
client_secret.json credentials.Once you have completed the onboarding process, verify API connectivity and loopback token health by running:
$ ggsrun status
With v5.3.21, ggsrun seamlessly executes inline or complex multi-line Google Apps Script code directly from your terminal or scripts without escaping issues:
--ss):$ ggsrun exe1 --ss "function test(e){
return 'ok ' + e.key;
}" -f "test" -v '{"key":"value1"}'
Piping script payloads via standard input completely bypasses shell quotation and variable expansion limitations, supporting arbitrary comments, regular expressions, and template literals:
$ cat << 'EOF' | ggsrun exe1 -f "test" -v '{"key":"value1"}'
function test(e) {
// Complex regular expressions, comments, and string escapes
const re = /^(https?:\/\/)?([\da-z\.-]+)\.([a-z\.]{2,6})([\/\w \.-]*)*\/?$/gi;
/* Multi-line comment:
Special characters: $ \ ' " ` \n