| What it does | CI/CD testing for AI voice agents that calls your agent and returns pass/fail results
| Evidence laid out, not a lone percentage.
|
|---|
| Stage | Not given
| Not given
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | Not given
| Not given
|
|---|
| Pricing model | Free
| Freemium
|
|---|
| Pricing | 100 free minutes. No credit card required.
| Free plan: 4 welcome credits (1 full report). Pro: $19/month for 300 credits/month. Team: $59/month for 1500 credits/month. Each report costs 4 credits.
|
|---|
| Key features | - Real phone calls from real carrier lines for end-to-end validation
- LLM-powered test agent that role-plays as a customer
- Plain English test case descriptions without scripted paths
- Simple two-call REST API integration
- Complete call transcripts in failure notifications for troubleshooting
- LLM as Judge for intelligent pass/fail evaluation
- No SDK or test case repository to manage
| - Cross-checks four independent detection models
- Deterministic evidence trump card (C2PA credentials, generator self-tags)
- Published weighted-vote rule (30% each for Sightengine, Winston AI, Was It AI; 10% for SAFE)
- Batch screening of up to 50 images at once
- Graded conclusions with disagreement reported
- Shareable report links and forensic PDF download
- Support for JPEG, PNG, WebP up to 12MB
- Generator-specific classification (Midjourney, DALL-E, Stable Diffusion, Flux, Imagen, Adobe Firefly, Ideogram)
- SHA-256 file hash recording for chain-of-custody
- Private uploads, shared only via report link
|
|---|
| What makes it different | - Real end-to-end validation with actual phone calls covering SIP integration, STT/TTS, and functional behavior
- Intelligent LLM testing that understands intent instead of matching keywords or regex
- Lightweight and unopinionated API design for fast integration
- Complete call transcripts included in test failures for immediate troubleshooting
| - Published weighted-vote rule with every vote shown by name instead of hidden percentages
- Reports disagreement honestly instead of blending signals into one number
- Automatic credit refund when an engine fails
- Full product access in free tier (not a limited teaser)
- Deterministic file evidence (C2PA, metadata) that outranks model votes
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 7 ● no change since last check | 40 ● no change since last check |
|---|
| Trust Flow (Majestic) | 3 | 13 |
|---|
| Citation Flow (Majestic) | 22 | 27 |
|---|
| Referring domains (Majestic) | 40 | 102 |
|---|
AI Image Detector