---
title: "Penetration testing for companies that run AI agents"
canonical: https://pharosone.ai/pentest
publisher: PharosOne (Agentic Mind, Inc.)
updated: 2026-10-22
license: "Open for reuse: crawling, copying, quoting, summarising and model training permitted without asking; citation of the title and canonical URL requested (see https://pharosone.ai/terms#reuse)"
cite_as: "\"Penetration testing for companies that run AI agents\". PharosOne, 2026-10-22. https://pharosone.ai/pentest"
---

# Penetration testing for companies that run AI agents

> Senior-led penetration testing of web apps, APIs, cloud, code and networks, with the AI agents on the stack checked in every engagement. From €9,500, report in week 3, retest included.

URL: https://pharosone.ai/pentest

Senior engineers test your web app, API, cloud, code or network at a fixed price. The AI agents reachable from that scope are checked in every engagement.

**From €9,500** · report in week 3 · one retest within 90 days included

A sample SOW for a fictional client. Scope: Web application and its API, 3 roles, about 75 endpoints; AI check: Included. Support agent, its tools and MCP server; Effort: 7 senior days, grey box; Report: Week 3, with a live debrief; Retest: One round within 90 days, closure letter; Price: €12,400, fixed, net of VAT.

## Every scope includes the AI exposure check.

Pick the scope your reviewer asks for. In each one the engineers list the agents and LLM features reachable from it, with what they read and which tools they call, and try to drive them through those inputs.

| Scope | What is in it | Effort | AI exposure check | From |
| --- | --- | --- | --- | --- |
| Web application and API | One application and its API, up to 3 roles, about 75 endpoints | 7+ senior days | Included | from €9,500 |
| Cloud environment | The cloud environment that hosts the application | 7+ senior days | Included | from €9,500 |
| Source code | The code behind the scope, findings confirmed on the running system | 7+ senior days | Included | from €9,500 |
| Internal network | Up to about 200 hosts and one AD domain, from an employee seat | 8–9 senior days | Included | from €12,000 |
| AI agents and LLM apps | One agent or LLM application with everything it touches: its tools, MCP servers, memory and knowledge base. Several agents or a multi-agent system are priced on the scoping call. | 10 senior days | The whole scope | from €15,000 |
| WAF, antivirus, EDR bypass | Whether the controls in front of the app or on the endpoint stop a real payload | 5–6 operator days | Included | from €9,500 |
| Red team | One objective, 6–8 weeks | ~25 operator days | Included | from €38,000 |

Prices net of VAT; each "from" price is the smallest standard scope, and the exact number is fixed in the SOW after the scoping call.

## Your agents are tested at two depths.

The support bot that reads your tickets is the entry point an attacker tries first. The exposure check covers it inside any scope. When the agent is the product, or it moves money or data, book the full agent pentest.

[The AI agent pentest, from €15,000](https://pharosone.ai/pricing#scopes)

## A report your auditor reads without a call.

Written for two readers: the engineer who fixes and the reviewer who signs. Retest is always included.

- **Executive summary**: The posture verdict, the conclusion and what is working, readable in three minutes.
- **Findings**: Each with a CVSS score, the affected assets, the evidence that reproduces it, the business impact and the fix.
- **Attack chain**: The path from the first foothold to the final impact, so you see which single fix breaks it.
- **Retest letter**: Every finding re-run after your fixes, closed with a letter the reviewer files.

One recent external perimeter test: 1 critical (remediated in two hours, verified by retest), 4 high, 14 medium, 12 low and below. 11 scanner findings discarded as false positives after manual checks.

[See a sample report](https://pharosone.ai/sample-report)

## The whole engagement fits in four weeks.

The engineer on the scoping call runs the test and writes your report, so nothing agreed on the call is lost in a handover.

1. **Scoping call** (Day 0): 30 minutes on the stack, the agents on it, the reviewer and the deadline.
2. **SOW** (Day 1–3): Fixed scope, fixed price, test window, rules of engagement.
3. **Testing** (Week 1–2): Senior engineers test the agreed scope, with the AI exposure check.
4. **Report** (Week 3): Findings, attack chain and standards mapping, with a live debrief.
5. **Retest** (Week 3–4): Every finding re-run once you are ready, within 90 days. Closure letter.

## The price is fixed before day one.

Net of VAT; each "from" price is the smallest standard scope. A second retest round, after the included one, costs 15% of the project price.

### Web apps · APIs · cloud · source code: from €9,500 (usually €12,000–18,000)

- 7+ senior days
- One app and its API, up to 3 roles, about 75 endpoints
- SOC 2 CC4.1 · PCI DSS 11.4 · ISO 27001 A.8 · DORA Art. 24–25

### AI agents and LLM applications: from €15,000 (one agent or LLM app with everything it touches)

- 10 senior days. Senior engineers direct our agents and re-check every result.
- Every bypass reproduced and confirmed by a person
- AIUC-1 · ISO 42001 · EU AI Act Art. 9 and 15 · OWASP Agentic Top 10

### Internal network: from €12,000 (usually €12,000–20,000)

- 8–9 senior days from an employee seat
- Up to about 200 hosts, one AD domain
- Written for PCI DSS 11.4 and DORA reviewers

Prices are net of VAT; each "from" price is the smallest standard scope. The exact number is fixed in the SOW after a 30-minute scoping call.

[All scopes and prices, red team and EDR bypass included](https://pharosone.ai/pricing)

## Senior engineers with a public record.

Only senior engineers test, and the engineer who scoped the project is the one who tests it. References available under NDA.

- **Published CVEs**: 22, by the team
- **Speaking**: Black Hat, speaker
- **Capture the flag**: DEF CON 28, IoT CTF winner
- **Certification**: OSEP, offensive lead
- **Open source**: PCF, and IoTSecFuzz, written by the offensive lead
- **Experience**: 15+ yrs, in cyber security

[About the team](https://pharosone.ai/about)

## Questions buyers ask about the pentest.

### Is a retest included in the price?

Yes. One retest within 90 days of the report is included and ends with a closure letter. A second retest round costs 15% of the project price.

### Will a SOC 2, PCI DSS or DORA reviewer accept the report?

The report states scope and methodology, rates every finding with CVSS, carries reproducible evidence, and the closure letter documents remediation. That is what SOC 2 CC4.1, PCI DSS 11.4, ISO 27001 A.8 and DORA Art. 24–25 reviewers look for.

### How long from the first call to the report?

The SOW is signed within 3 days of the scoping call, testing runs in weeks 1 and 2, and the report arrives in week 3.

### How is this different from the €3,900 AI-run pentest?

Both use PharosOne's AI agents. The AI-run pentest is autonomous: our agents run the whole test against one agent, and an engineer reviews the findings and signs the report in 5 working days. The human pentest is engineer-led: senior engineers direct the agents, validate every result and re-test it by hand over 7 or more days, with a full report and a retest. The €3,900 is credited in full to a project ordered within 90 days.

### Are the agents on my stack part of a web or network pentest?

Yes. Every engagement includes the AI exposure check of the agents reachable from the tested scope. When the agent is the product, book the AI agents scope, from €15,000.

## Book a 30-minute scoping call

With the engineer who runs the test. You leave the call with a fixed scope, and the SOW follows within three days.

[Book a scoping call](https://pharosone.ai/book) · [See a sample report](https://pharosone.ai/sample-report)

- One field to start: work email. The calendar asks the rest.
- Bring the reviewer's name and the audit date.
- Prefer email? audit@pharosone.ai

Source: PharosOne, "Penetration testing for companies that run AI agents", https://pharosone.ai/pentest, updated 2026-10-22.

---

PharosOne is a product of Agentic Mind, Inc., an independent AI agent security company. Site map for agents: https://pharosone.ai/llms.txt