Hear every voice in your language
Live AI dubbing and bilingual subtitles for any video. Free, open source and private by design.
A Claude Code plugin that warns you when a conversation gets too long and shows token usage per model after every answer.
PluginsJavaScriptGitHub @FabioLima77OnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Claude Code plugin: warns about long, expensive chats and shows token usage per model after every answer.
Translation pending
Taken from the repository README
Loading the full README.
Read the full README ↗Spend fewer tokens in Claude Code. Economy Tokens warns you when a conversation gets too long (and expensive), and shows how many tokens each model used after every answer.
Built by FabioLima77. Português: see README.pt-BR.md.
Claude reads the whole conversation again on every answer. A chat with 300,000 tokens costs far more per message than a fresh one. Economy Tokens makes that cost visible and nudges you to start fresh at the right moment.
| Part | When | What it does |
|---|---|---|
| Model line | Start of each answer | Opens every reply with a line naming the model that is answering, for example [Haiku 5.5 · principal], so you always know which model you are paying for. |
| Long-conversation warning | Before each prompt | Warns and asks Claude to suggest a new conversation with a summary, once the chat passes your limit. |
| Token footer | After each answer | Shows tokens per model: main model, subagents and advisor. |
economy-tokens skill | When you ask about saving tokens | Explains the habits that keep costs low. |
Everything runs on your computer. Nothing is sent anywhere.
After this repository is public, run these two commands in a terminal, one at a time:
claude plugin marketplace add FabioLima77/economy-tokens-claude-code
claude plugin install economy-tokens@economy-tokens-marketplace
To install from a folder on your computer instead:
claude plugin marketplace add "C:\path\to\economy-tokens-claude-code"
Then start a new conversation so the hooks load.
| Variable | Default | Meaning |
|---|---|---|
ECONOMY_TOKENS_LIMIT | 300000 | Tokens read per answer before the long-conversation warning appears. |
Set it in the env block of ~/.claude/settings.json, for example "ECONOMY_TOKENS_LIMIT": "170000". |
To choose which model runs the main chat, the subagents and the advisor, see docs/model-setup.md.
Pause it:
claude plugin disable economy-tokens@economy-tokens-marketplace
Turn it back on:
claude plugin enable economy-tokens@economy-tokens-marketplace
Remove it:
claude plugin uninstall economy-tokens@economy-tokens-marketplace
/clear) and paste the summary.economy-tokens-claude-code/
├── .claude-plugin/
│ ├── plugin.json # plugin identity
│ └── marketplace.json # lets you install with "claude plugin marketplace add"
├── hooks/
│ ├── hooks.json # connects the two hooks to the right moments
│ ├── warn-long-conversation.js
│ └── token-footer.js
├── skills/
│ └── economy-tokens/SKILL.md # habits to save tokens
└── docs/
├── model-setup.md
└── index.html # presentation page
If this saves you money, you can support the project with a donation on GitHub Sponsors. Thank you!
MIT. See LICENSE.