opencode-quota

Quota and token tracking for OpenCode providers via Toasts and Commands with no LLM calls.

What It Does

Quota Toasts - See your remaining quota at a glance after each assistant response.

Token Report Commands - Track token usage and estimated costs across sessions.

Installation

Add to your opencode.json:

{
  "plugin": ["@slkiser/opencode-quota"]
}

Quick Start

Enable the providers you use:

{
  "experimental": {
    "quotaToast": {
      "enabledProviders": ["copilot", "openai", "google-antigravity"],
    },
  },
}

That's it. Toasts appear automatically after main agent responses.

Commands

Command	Description
/quota	Show current quota (verbose)
/quota_today	Token + cost report (today, local timezone)
/quota_daily	Token + cost report (last 24 hours)
/quota_weekly	Token + cost report (last 7 days)
/quota_monthly	Token + cost report (last 30 days)
/quota_all	Token + cost report (all history)
/quota_chat	Token + cost report (current session only)
/quota_status	Diagnostics (config, paths, pricing)

Supported Providers

Provider	Config id	Notes
GitHub Copilot	copilot	Uses OpenCode auth*
OpenAI (Plus/Pro)	openai	Uses OpenCode auth
Firmware AI	firmware	Uses OpenCode auth or API key
Google Antigravity	google-antigravity	Multi-account via opencode-antigravity-auth

Firmware AI Setup

Firmware works automatically if OpenCode has Firmware configured. Alternatively, you can provide an API key in your opencode.json:

{
  "provider": {
    "firmware": {
      "options": {
        "apiKey": "{env:FIRMWARE_API_KEY}",
      },
    },
  },
  "experimental": {
    "quotaToast": {
      "enabledProviders": ["firmware"],
    },
  },
}

The apiKey field supports the {env:VAR_NAME} syntax to reference environment variables, or you can provide the key directly.

GitHub Copilot Setup (optional)

Copilot works with no extra setup as long as OpenCode already has Copilot configured and logged in.

Optional: if Copilot quota does not show up (or you want more reliable quota reporting), you can provide a fine-grained PAT so the plugin can use GitHub's public billing API:

1. Create a fine-grained PAT at GitHub with Account permissions > Plan > Read
2. Create ~/.config/opencode/copilot-quota-token.json:

{
  "token": "github_pat_...",
  "username": "your-username",
  "tier": "pro"
}

Tier options: free, pro, pro+, business, enterprise

* The plugin reads Copilot auth from OpenCode. The PAT file is only a fallback for reliability.

Configuration Reference

All options go under experimental.quotaToast in opencode.json:

Option	Default	Description
enabled	true	Enable/disable plugin
enableToast	true	Show popup toasts
enabledProviders	[]	Provider ids to query
minIntervalMs	300000	Min ms between fetches (5 min)
toastDurationMs	9000	Toast display time
onlyCurrentModel	false	Only show current model's quota
googleModels	["CLAUDE"]	Google models: CLAUDE, G3PRO, G3FLASH
debug	false	Show debug info in toasts

Troubleshooting

Toast not appearing? Run /quota_status to check config and provider availability.

License

MIT