LaunchRoundsSubmit Software

PerfAI: Find and fix broken app permissions

PerfAI is an application security platform that maps, tests and remediates access control permissions in apps built with AI coding tools.

What is PerfAI?

PerfAI is an application security platform focused on access control in apps produced with AI coding tools. It maps the permissions across an app's UIs, APIs, objects and roles, then tests them to surface broken or risky ones. The vendor illustrates the scale with an average app of 80 UIs, 200 APIs and 100 objects across 6 roles, which it counts as 2,280 permissions. Beyond detection, the platform offers a Fix Agent that produces a vetted code patch for confirmed vulnerabilities and delivers it as a pull request or pushes it into AI coding tools. Plans differ by credits, app size, endpoint counts, testing frequency and which coding tools the Fix Agent supports, with an Enterprise tier covering portfolio-wide apps, multi-tenancy and on-premises deployment.

Who is PerfAI for?

  • Solo founders shipping AI-built apps to production
  • Startups facing a first compliance request
  • Security teams and MSSPs managing many AI-built apps

How much does PerfAI cost?

PerfAI starts at $49/mo. Always check the vendor’s own pricing page before you buy.

PlanPriceWhat you get
Free$0500 credits/mo, 1 Small app, 50 UI & API endpoints, top 5 findings in full detail, community support
Pro$49/mo per app1,000 credits/mo, 1 Medium app, 100 UI & API endpoints, Fix Agent for Replit / Lovable / Base44, Perfai Chatbot, email support
Startup$299/mo per app3,000 credits/mo, 1 Large app, 150 UI & API endpoints, weekly autonomous testing, Fix Agent for Cursor / Claude Code / Copilot / Devin, compliance-ready reports, Slack & Teams, priority email support
EnterpriseCustomPortfolio-wide apps, custom collaborators and endpoints, private agents, on-prem deployment, CI/CD gating, multi-tenancy, API access, dedicated contact

Is PerfAI worth it?

A focused access control testing service for teams building apps with AI coding tools, with a usable free tier and auto-fix on paid plans

Where PerfAI is strong

  • Maps and tests permissions across UIs, APIs, objects and roles
  • Fix Agent generates patches and pull requests for confirmed issues
  • Free tier with 500 credits per month and no credit card

Where it falls short

  • Free plan shows only the top 5 findings in full detail
  • No independent user reviews or third-party validation were found in the source

How do you get started with PerfAI?

  1. 01Sign in with Google or email and start on the free plan
  2. 02Paste your app URL so Perfai can learn the app and map its permissions
  3. 03Review the permission issues found and apply fixes where available

PerfAI: frequently asked questions

Is there a free plan?

Yes. The Free plan costs $0 and includes 500 credits per month, one small app, 50 UI and API endpoints, the Vision and Security Agents, and the top 5 findings in full detail with a teaser for locked findings.

What does the Fix Agent do?

It generates a vetted code patch for confirmed vulnerabilities and ships it as a pull request, or pushes it directly into AI coding tools such as Cursor, Claude Code, GitHub Copilot, Replit and Windsurf. It is not included on the Free plan.

Which AI coding platforms are covered?

The source lists Cursor, Bolt, v0, Replit, Windsurf, Claude Code, GitHub Copilot, Devin, Codeium, Aider, StackBlitz and Vercel v0.

How does it differ from a pentest or DAST scan?

PerfAI describes itself as continuous and agentic, with autonomous agents discovering, exploiting and remediating around the clock, in contrast to point-in-time pentests and static scans.

PerfAI alternatives

The closest tools in our directory, matched on category, focus areas and audience.

PerfAI compared with its alternatives
ToolBest forFree planStarting priceFocus areas
PerfAITeams securing access control in AI-built appsYes$49/mo—
SnykDevelopers and teams adding security checks to their build processYes$25/mo—
SuperagentTeams wanting continuous security checks on code, agents and infrastructureYes$499/mo—
AegisTeams wanting a free starting point for application threat blockingYes——
AsymptoteSecurity teams governing coding agentsYes——
CorgeaSecurity and engineering teams fixing exploitable code risk fasterYes——
Polymer Runtime Data SecuritySecurity and compliance teams securing data in AI and SaaS toolsNo$5/mo—
EthiackTeams wanting continuous, proof-backed validation of exploitable security risksNo——
Upwind SecurityEnterprise cloud security teamsNo——
CodeAnt AITeams wanting exploit-proven security findings across code and cloudUnknown——
Pentest CopilotContinuous autonomous pentesting of apps, networks and codeUnknown——