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.
npx agentmods add commands/stevegustafson32/google-multi-account/setupgit clone --depth 1 https://github.com/stevegustafson32/google-multi-accountWrote 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/commands/stevegustafson32/google-multi-account/setup)<a href="https://agentmods.dev/commands/stevegustafson32/google-multi-account/setup"><img src="https://agentmods.dev/badge/commands/stevegustafson32/google-multi-account/setup.svg" alt="Measured on agentmods" 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 | $0.00012 | $0.03111 |
| Opus 5 | $0.00006 | $0.01555 |
| Sonnet 5 | $0.00002 | $0.00622 |
| Haiku 4.5 | $0.00001 | $0.00311 |
Grade A, and why
setup 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.
How it starts
The opening of the file, as written. The whole thing — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Google Multi-Account MCP — Automated Setup
Guide the user through setting up the Google Multi-Account MCP plugin. This command handles everything: copying server files, installing dependencies, registering with Claude Desktop, and authenticating Google accounts.
CRITICAL: Pre-baked OAuth Credentials
This plugin ships with pre-baked OAuth credentials. Users do NOT need to create a GCP project, enable APIs, or generate their own OAuth client. The setup-tokens.js script uses shared credentials automatically.
NEVER walk users through GCP project setup unless they explicitly ask to use their own credentials.
CRITICAL LESSONS (from real-world installation failures)
- Cowork IGNORES plugin
.mcp.jsonfiles. MCP servers MUST be registered in~/Library/Application Support/Claude/claude_desktop_config.jsonto connect. - NVM/volta/fnm node is invisible to Cowork. Never use bare
nodeas the command. Use/bin/bashwith thestart.shwrapper script that auto-discovers node. - The MCP server must NEVER call
process.exit()on config errors. Cowork marks crashed servers as permanently dead and won't retry. The server starts with zero tools instead. - Plugin zip extraction is unreliable. Server files may not land in the plugin directory.
Always copy server source to
~/.google-multi-mcp/server/as the canonical location. - Paths with spaces break things. The plugin install path contains "Application Support"
and deep UUID directories. Always use
~/.google-multi-mcp/as the clean install target.
Step 0: Detect the platform FIRST
Every path in this document differs by OS. Determine which one you're on before running anything, and use only that column.
uname -s 2>/dev/null || echo "Windows"
| macOS | Windows | |
|---|---|---|
| Install dir | ~/.google-multi-mcp/ |
%USERPROFILE%\.google-multi-mcp\ |
| Launcher | start.sh |
start.cmd |
| Desktop config | ~/Library/Application Support/Claude/claude_desktop_config.json |
%APPDATA%\Claude\claude_desktop_config.json |
MCP command |
/bin/bash |
cmd |
MCP args |
["/Users/<you>/.google-multi-mcp/start.sh"] |
["/c", "C:\\Users\\<you>\\.google-multi-mcp\\start.cmd"] |
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.
- 3d ago First seen · 306 lines · 12 tokens per session scan A 84e3c0077cee
setup is a command published in the GitHub repository stevegustafson32/google-multi-account (2 stars, last pushed 1mo ago), licensed MIT. It adds 12 tokens to every session and 3,111 once invoked, about $0.0001 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.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.