| What it does | Catch where AI agents get stuck on your API or MCP server, rank problems by impact, and draft fixes as pull requests
| Open-source runtime for computer-use agents
|
|---|
| Stage | Not given
| Startup
|
|---|
| Founded | Not given
| 2026
|
|---|
| Based in | US
| Not given
|
|---|
| Pricing model | Freemium
| Open source
|
|---|
| Pricing | Open source: free. Team: $199/month with 14-day free trial (1 API or MCP service, 25,000 reports/month, 15 coding-agent runs/month). Business: $799/month (up to 3 services, 250,000 reports/month, 60 coding-agent runs/month). Extra coding-agent runs: $6 each.
| Not given
|
|---|
| Key features | - Session replay for MCP servers to identify failed agent sessions
- Clustering reports by agent goals and workflows
- Priority ranking weighted by frequency, failed workflows, and affected accounts
- Claude Code generates pull requests with tests
- Review queue with diff, tests, agent reports, and evidence
- Isolated git worktree prevents unauthorized code push
- Schema validation and rate limiting for feedback
- Open Agent Feedback Protocol with TypeScript and Python SDKs
- GitHub integration for draft PR creation
| - Works with any OpenAI-compatible model
- Ready-wait mechanism instead of fixed sleeps
- Batches actions and bounds model calls
- Desktop pause, resume and fork with memory
- MCP server for Claude Code and Cursor
- Graded rollouts for computer-use RL
- Works with full desktops
- Supports multiple sandbox providers
|
|---|
| What makes it different | - Only AI agent failures on successful HTTP responses get detected and ranked
- Automated pull request drafts with your approval required before shipping
- Session replay captures failures even when agents don't report them
- Pay only for coding-agent runs, not feedback submission
- Open protocol and SDKs can be self-hosted at no cost
| - 1.48x faster than Anthropic's computer-use reference loop on the same model
- 19 points more success than OSWorld's Qwen agent at same speed
- Reduces rollout time by 40-53% by eliminating unnecessary waiting
- Published benchmarks run same model on same desktops with reproducible results
- Works with any model and sandbox provider
|
| Community votes | 0 | 0 |
| DR (Domain Rating by Ahrefs) | 1 ● no change since last check | Not checked yet ● no change since last check |
| Trust Flow (Majestic) | 1 | 0 |
| Citation Flow (Majestic) | 14 | 0 |
| Referring domains (Majestic) | 4 | 0 |