Practical overview
When it fits
Use Tokscale when coding-agent usage is distributed across several supported local data sources and you need one place to inspect token consumption and costs.
Who it is for
- Developers using multiple AI coding agents.
- Teams or individuals reviewing token consumption and costs across coding workflows.
- Builders extending a Rust-based usage tracker with additional coding-agent data sources.
Documented uses
- Aggregate token usage from supported coding-agent session data.
- Review token costs across different agents or models.
- Compare usage patterns between coding-agent workflows.
- Visualize local agent activity through CLI and dashboard views.
Requirements & limits
The project repository is public and licensed under MIT. It is presented as a CLI and desktop-oriented visualization tool; no verified pricing or account requirement is provided.
- Supported integrations read different local databases, session files, transcripts or exports.
- Cursor usage is handled through an API export cached at ~/.config/tokscale/cursor-cache/usage*.csv.
- The project is implemented in Rust.
- The repository identifies support for a broad set of coding-agent ecosystems, including Claude Code, Codex CLI, Gemini CLI, Cursor and OpenCode.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Aggregate local coding-agent usage
Set up Tokscale to read and visualize usage records from supported AI coding agents.
Before you start
- Tokscale repository and CLI access.
- At least one supported coding agent with usage data in the location listed by the project.
- Rust-based Tokscale tool or its published CLI distribution.
Steps
Choose a supported client such as Claude Code, Codex CLI, Gemini CLI, Cursor or OpenCode.
Confirm that the client’s listed local database, session, transcript or exported CSV location contains usage data.
Run Tokscale from the terminal to aggregate the supported records.
Review the resulting CLI or visualization dashboard views for token usage and costs.
Expected outcome
A consolidated view of token consumption and costs for the supported local agent records.
Guide limitations
- Each integration uses a different data source and path.
- Cursor relies on an API export cached by Tokscale rather than a ~/.cursor directory.
- The repository documentation does not establish that every listed client produces data in every environment.
Compare options for AI coding assistance
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Tokscale | Aggregate token usage from supported coding-agent session data. | CLI, Desktop | The project repository is public and licensed under MIT. It is presented as a CLI and desktop-oriented visualization tool; no verified pricing or account requirement is provided. |
| Harness Score | Measure the presence and structure of agent guidance, skills, hooks, feedback tools and safeguards. | CLI | The project is documented as a CLI and is MIT-licensed. The README shows invocation through `npx harness-score` and references a GitHub Action. |
| Agenttrail | See which project components and files are changing. | CLI, Desktop | Kitchen setup requires Node.js 20 or later, a browser with WebGL and a local project folder. The README describes Kitchen as a public experimental preview. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 2 October 2026
- Original website title
- junhoyeo/tokscale · Official homepage title · retained in its source language