---
title: "token-calculator"
description: "Analyze token usage, estimate API costs, compare AI models, and assess file sizes for context window optimization. A toolkit for understanding and managing AI resource consumption."
canonical: https://agentpluginsdirectory.com/plugins/token-calculator
last-updated: 2026-09-21
---

# token-calculator
Analyze token usage, estimate API costs, compare AI models, and assess file sizes for context window optimization. A toolkit for understanding and managing AI resource consumption.
- Slug: token-calculator
- Publisher: Pablo
- Repository: https://github.com/novanoticia/token-calculator-suite
- Manifest: plugin.json
- Version: 2.0.0
- Category (editorial): agent-tooling
- Skills: 4 (file-analysis, model-comparison, project-cost, token-analysis)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-03-25
- Repository last pushed: 2026-08-07
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/token-calculator
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What token-calculator does, in the publisher's words

> Conforms to Agent Plugins 1.0.0: the > portable packaging format from the Agentic AI Foundation (OpenAI, Amazon, Microsoft, > Cursor and Vercel, with Google as core maintainer). The repository root is the plugin > root: it carries the portable plugin.json manifest alongside.claude-plugin/ and > the four skills in skills/token-analysis/, skills/file-analysis/, > skills/project-cost/ and skills/model-comparison/, so any conformant client > discovers them. > > Works in ChatGPT. The skills are plain text, instructions and criteria, no local > execution, so they install by switching Work on in the ChatGPT selector and > adding them from Plugins, by name or by this repository's URL. They behave the > same as in Claude. Their frontmatter validates against the > closed Agent Skills field set, which is what > ChatGPT, claude.ai and the Skills API require to accept an upload: an extra key there > is not ignored, it fails hard. Also available on the free plan, with usage limits.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/novanoticia/token-calculator-suite/HEAD/README.md

## Skills

- file-analysis: Estimate token consumption of files before uploading to Claude. Use when the user wants to know how much context a file will consume, whether a file fits in the context window, or needs to plan which files to include in a conversation.
- model-comparison: Compare AI models by capability, cost, context window, and best use cases. Use when the user asks which model to use, compares Claude vs GPT vs Gemini vs DeepSeek, or needs help choosing the right model for a task.
- project-cost: Estimate total API costs for projects and workflows. Use when the user wants to budget for API usage, plan a project involving multiple Claude calls, or understand the cost of automating a workflow.
- token-analysis: Analyze text for token count estimation and API cost calculation. Use when the user asks about tokens, costs, context window usage, or wants to optimize prompt length.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.