What is Agent-as-a-Service pentesting?

Agent-as-a-Service (AaaS) pentesting lets you chat with autonomous pentesting agents that scan your application on demand. Instead of waiting weeks for a manual engagement, you talk to specialized agents — pen-scout (recon and surface mapping), pen-recon (deeper enumeration), pen-triage (validates and prioritizes findings), pen-fixer (remediation guidance), and pen-compliance (OWASP/standards mapping). Findings are validated with proof-of-concept and re-test, not just flagged. You start with 150 free credits, no credit card.

What are credits and how do they work?

Credits are how you run agent scans. Each scan or agent action consumes credits based on its depth. New accounts get 150 free credits with no card required. After that you can buy one-time credit packs ($29, $79, or $199) for pay-as-you-go use, or subscribe to a monthly tier ($49, $149, or $399/mo) for continuous, on-demand pentesting. The agents pen-scout, pen-recon, pen-triage, pen-fixer and pen-compliance all draw from the same credit balance.

Yes. Every new account gets 150 free credits with no credit card required — enough to chat with the pentesting agents and run real scans against your app. There is also a free security headers scan at sable.somoswilab.com/free-scan and a sample report at sable.somoswilab.com/sample-report. The free tier runs on OpenRouter models so you can evaluate the autonomous agents before paying.

What is penetration testing for startups?

Penetration testing (pentesting) is a simulated cyberattack against your application to find security vulnerabilities before real attackers do. For startups, we focus on the issues that matter most at your stage: authentication flaws, data exposure, API security, and common mistakes in modern stacks like Next.js, Supabase, and Firebase.

How much does a pentest cost?

Traditional pentests cost $10,000-$50,000+. SableOffensive starts at $29 for a Pre-Launch Check covering OWASP Top 10 and secrets detection. Founder Shield ($79) adds IDOR testing, auth bypass, and a debrief call. Scale Secure ($199) is a full-scope assessment. Every plan includes a professional report with remediation steps.

How long does a security scan take?

Pre-Launch Check reports are delivered within 24-48 hours. Founder Shield and Scale Secure may take 2-3 business days depending on the complexity of your application.

What do I need to provide?

At minimum, just your application URL. For more comprehensive testing, we may ask for staging credentials, API documentation, or GitHub repository access. We sign NDAs for all engagements.

What is OWASP Top 10?

OWASP Top 10 is the industry standard list of the most critical web application security risks. It includes injection attacks, broken authentication, cross-site scripting (XSS), server-side request forgery (SSRF), and security misconfigurations. Every SableOffensive assessment tests against the full OWASP Top 10.

Do you test AI-generated code?

Yes. Code generated by AI tools like Cursor, GitHub Copilot, and v0 often contains subtle security issues: hardcoded secrets, missing input validation, insecure API patterns, and overly permissive access controls. We have specific testing procedures for AI-generated codebases.

How do you secure Supabase and Firebase apps?

For Supabase, we audit Row Level Security (RLS) policies, test for direct table access, and check for exposed service keys. For Firebase, we review security rules, test Firestore/RTDB access patterns, and check Cloud Functions for vulnerabilities.

What if you find zero vulnerabilities?

50% money back guarantee. If our scan finds zero security issues, you get half your money back.

Is there a free pentesting option?

Yes. SableOffensive offers a free security headers scan at sable.somoswilab.com/free-scan. It instantly checks your website for 8 critical security headers (HSTS, CSP, X-Frame-Options, and more) and gives you an A-F grade with copy-paste fixes. No signup or payment required.

Can I get a free vulnerability scan?

Our free security headers check scans your website instantly and grades your security posture. For a deeper free assessment, contact us — we occasionally offer complimentary scans for early-stage startups and open source projects.

PraisonAI Auth Bypass Was Scanned 4 Hours After Disclosure

What Happened

On May 14, 2026, an authentication bypass vulnerability in PraisonAI — a multi-agent AI framework used to deploy autonomous agents for complex task execution — was publicly disclosed as CVE-2026-44338. Within four hours, security firm Sysdig observed active scanning targeting exposed PraisonAI instances. SecurityWeek reported that exploitation attempts began almost immediately after the CVE dropped, with attackers probing for internet-facing deployments running the default-insecure configuration. CSO Online confirmed the near-instant weaponization timeline, underscoring a growing trend: AI infrastructure is now a first-day exploitation target.

Technical Analysis

PraisonAI is a multi-agent teams system that allows organizations to orchestrate autonomous AI agents for complex workflows. The vulnerability exists in a legacy Flask API server that PraisonAI ships and generates with authentication disabled by default. From version 2.5.6 through before version 4.6.34, any caller who can reach the server can access the /agents endpoint and trigger the configured agents.yaml workflow through /chat — without providing any token. The GitLab advisory confirms that the legacy API server exposes full agent workflow execution to unauthenticated requests. Sysdig's analysis notes that this isn't just a theoretical risk — the exposed endpoints allow attackers to hijack configured AI agents, potentially accessing the tools, data, and API keys those agents are authorized to use. The attack surface includes any PraisonAI deployment where the legacy Flask server is reachable over the network, which in practice means any internet-facing instance with default settings.

Who's Affected

Any organization running PraisonAI versions 2.5.6 to before 4.6.34 with the legacy API server enabled is vulnerable. The framework has seen rapid adoption as enterprises experiment with multi-agent AI workflows — automating everything from data analysis to code generation to customer support. OpenCVE data shows the vulnerability affects the PyPI package, meaning any pip-installed deployment is in scope. The real danger is lateral: compromised AI agents often have access to internal APIs, databases, cloud credentials, and third-party services. An attacker who bypasses authentication on the PraisonAI server doesn't just get a chat interface — they get the full execution context of every configured agent, including whatever tools and permissions those agents were granted. This makes CVE-2026-44338 a potential gateway to cloud infrastructure, source code repositories, and production databases.

How to Protect Yourself

1. Upgrade to PraisonAI 4.6.34 or later immediately. The patched version addresses the authentication bypass. If you're running any version from 2.5.6 up to (but not including) 4.6.34, you are vulnerable.

2. Disable the legacy Flask API server if not actively needed. If your deployment doesn't require the legacy API, disable it entirely. Attackers can't exploit what isn't running.

3. Enforce authentication on all API endpoints. Even if you're running a patched version, verify that authentication is explicitly enabled — not just present in code, but actively required. Test with unauthenticated requests to confirm.

4. Network-segment your AI agent infrastructure. PraisonAI servers should not be directly internet-facing. Place them behind a reverse proxy with proper auth, or restrict access to internal networks and VPNs.

5. Audit agent permissions and tool access. Assume breach: review what tools, APIs, and credentials your configured agents can access. Apply least-privilege principles. If an agent doesn't need production database access, revoke it now.

6. Monitor for scanning activity. The 4-hour window between disclosure and scanning means your logs may already contain reconnaissance hits. Look for requests to /agents and /chat endpoints from unfamiliar IPs.

The Sable Angle

CVE-2026-44338 is a textbook example of what we call the AI agent attack surface explosion. Organizations are deploying autonomous agents at scale — connecting them to production systems, cloud APIs, and sensitive data — without applying the same security rigor they'd demand of any other internet-facing service. The result: frameworks shipping with authentication disabled by default, agents running with excessive permissions, and zero network segmentation. At Sable, we've been tracking this trend closely. Our offensive research consistently finds that AI agent deployments are among the weakest links in modern infrastructure. See our analysis of Hugging Face's double-CVE for another case where AI tooling introduced critical, overlooked risk.

The PraisonAI disclosure should be a wake-up call: your AI agents are only as secure as the least-protected endpoint they run on. If you're deploying multi-agent frameworks in production, you need offensive testing that specifically targets agent orchestration layers — not just traditional web app scans. Talk to our team about AI-agent-specific security assessments before your agents become someone else's attack path.