AI Workflow QA For Agencies

Break your AI workflows before customers do.

AgentPrey turns client intake rules into repeatable scenarios, evaluates AI receptionist behavior, and produces report-ready evidence your agency can hand to clients before launch.

Request AuditSample Report
Manual transcript QA
Webhook endpoint runs
White-label reports

Product

A QA workspace for AI intake.

The MVP focuses on agencies shipping local-service AI receptionists and lead-intake workflows. The legacy CLI scanner remains supported in docs and uploaded scan reports.

Model the client workflow

Capture services, service area, booking rules, escalation policy, pricing constraints, and required lead fields.

Generate scenario coverage

Start from local-service templates for pool, HVAC, plumbing, and generic home-services receptionists.

Evaluate transcripts or endpoints

Grade manual transcript paste-ins or run bounded webhook tests against staging agents.

Share report evidence

Turn case results, transcripts, failures, and recommendations into public-by-link client reports.

Evidence

Reports built for client delivery.

Every run stores the transcript, scenario, failure evidence, risk score, and suggested fix. Share links can show the QA artifact without exposing your API key or webhook secrets.

View Sample

BrightPool Service Co.

QA Score 68

Fail
Quoted a specific price despite quote-required policy.
Skipped phone and address before treating the lead as ready.
Refused instruction override without leaking hidden prompt text.

Pricing

Audit first. Automate after.

Start with a paid launch audit for immediate proof, then keep recurring coverage in the Agency Pro beta workspace.

Legacy CLI

$0open source

Local prompt-injection scanner, CLI reports, and the legacy upload/share path remain supported.

CLI Docs
Fastest path to cash

Launch QA Audit

$750flat per workflow

One AI receptionist or lead-intake workflow tested across 15 launch scenarios, with a client-ready evidence report.

  • Workflow profile and rule capture
  • 15 adversarial and policy scenarios
  • Transcript, call, or endpoint evaluation
  • Shareable QA report with risk score
  • Prioritized fixes and launch/no-launch summary
Request Launch Audit
Follow-on coverage

Agency Pro Beta

$24per month
$240 per year
  • API-key QA workspace beta
  • Client workflow profiles
  • Scenario generation for local-service receptionists
  • Manual transcript evaluation
  • Webhook endpoint test runs
  • Public-by-link QA reports
  • Legacy CLI upload and scan reports
Start MonthlyStart Annual