| What it does | A fast, lean coding agent for your terminal
| Up to 3x more work with the same tokens on Claude Code and Codex
|
|---|
| Stage | Not given
| Not given
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | Not given
| BR
|
|---|
| Pricing model | Open source
| Freemium
|
|---|
| Pricing | Not given
| Basic FREE: always free with measurement and monitoring tools. Pró: US$9/month with 7 days free trial, includes Delta Read, Code Finder, Terminal Guide, Stop Loss, Budget Cap, and Git Admin. Ultra: US$12/month with 7 days free trial, adds Cheap Swarm and security features. GIGA UNLIMITED: US$27/mont
|
|---|
| Key features | - Starts in 13ms and uses only 17MB peak memory
- Works with any model: local or hosted APIs
- Reads Claude Code and Codex settings automatically
- Background runs on separate branches
- Verifies work by running project checks
- Undo function to revert changes
- Sandbox environment for safe command execution
- Spend limits and usage tracking
- MCP server support and plugins
- Scheduled runs
| - Deep Scan: measures token consumption by analyzing conversation history
- Smart Compress: compresses requests without calling the model
- Delta Read: reduces rereading of context across turns
- Cheap Swarm: routes lightweight tasks to cheaper models with 26 task classes
- Secret Lock: masks passwords and keys before they enter conversations
- Code Deck: 4 concurrent agent slots with 17+ CLI options
- KanBan Sync: automatic task board synced with agent planning
- HarnessUp Flow: create agentic workflows on infinite canvas
- Board: infinite canvas for notes, images, and columns
- Stop Loss: budget cap to prevent exceeding spending limits
|
|---|
| What makes it different | - Fastest coding agent: 1.5ms CPU per turn vs 4.2-145ms for other agents
- Lowest memory footprint at 17MB peak vs up to 960MB for competitors
- Most efficient token usage on real tasks
- Single 20MB portable binary requiring no installation
| - Measures token waste with your own data before charging anything
- Reverses AI damage with one button undo functionality
- No login required to start measuring token usage
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 2 ▲ +2 (up) since last check | Not checked yet ● no change since last check |
|---|
| Trust Flow (Majestic) | 0 | 0 |
|---|
| Citation Flow (Majestic) | 8 | 0 |
|---|
| Referring domains (Majestic) | 1 | 0 |
|---|
Chanto
HarnessUp