Catalogue · 100 tools · rev 2026-11-08
AI pentesting tools compared on what they publish
Autonomous pentest platforms, tools that red-team LLM applications, and open-source AI agents. What each one tests, where it runs, what it costs and how its findings are checked, from the maker’s own pages, linked and dated.
The catalogue holds 45 commercial platforms, 35 AI red-teaming tools and 20 open-source agents. 32 of the 45 platforms publish a price; the open-source agents are free to run, but you pay for the model. Choose by target first: web apps and APIs, networks and Active Directory, or an LLM application.
All 100 tools
Grouped by lane. Platforms are ordered by how much they publish; open-source entries by GitHub stars on the check date. Select a name for the full entry, its sources and what it says about reviewing findings.
| Tool | Type | Tests | Runs as | Price, or stars and licence |
|---|---|---|---|---|
| Platform · 45 · full entries | ||||
| Aikido AI PentestAikido Security · aikido.dev | Commercial | Web apps, APIs, Networks, Mobile, LLM apps, AI agents, Code | SaaSSelf-hosted | Typical pentest $4,000 / €3,500 / £3,000 per assessment |
| Burp ATPortSwigger · portswigger.net | Commercial | Web apps, APIs | Desktop | Public beta: $19 / $89 / $249 per user per month (Standard / Plus / Power) |
| EthiackEthiack, S.A. · ethiack.com | Commercial | Web apps, APIs, Networks, Cloud, Mobile | SaaS | Pentest from €3,000 per test |
| HacktronHacktron AI Inc. · hacktron.ai | Commercial | Code, Web apps | SaaSSelf-hosted | Whitebox $50 to $25,000 per pentest (about $500 per 100,000 lines of code) |
| Intruder AI PentestingIntruder Systems Ltd · intruder.io | Commercial | Web apps, APIs | SaaS | $3,500 per test |
| NeoProjectDiscovery · projectdiscovery.io | Commercial | Web apps, APIs, Code, Networks, Cloud, Mobile | SaaSSelf-hosted | Free $0 |
| PenligentFuture Share LLC · penligent.ai | Commercial | Web apps, APIs, Networks | Self-hostedDesktop | Free $0 |
| Pentest-Tools.com AI Pentests (Specter)Pentest-Tools.com · pentest-tools.com | Commercial | Web apps | SaaS | $3,000 per application (AI Pentest, incl. 2 free retests) |
| PentiPenti (formerly Securily) · penti.ai | Commercial | Web apps, APIs, Networks, Active Directory, Cloud, Mobile, LLM apps | SaaSSelf-hosted | Starter $20/mo |
| PurpleRidge SecurityRidge Security · purpleridge.ai | Commercial | Web apps, Cloud | SaaS | Credit packs: Starter $399 (100 credits), Essential $1,699 (500 credits), Advanced $2,999 (1,000 credits) |
| Siemba Autonomous PentestingSiemba · siemba.io | Commercial | Web apps, APIs, LLM apps, AI agents | SaaSMCP | Autonomous Pentesting from $500/month (25 test runs/month) |
| TurboPentestIntegSec · turbopentest.com | Commercial | Web apps, APIs, Networks, Code, Cloud, LLM apps | SaaS | $99 per pentest (Audit-Ready: 1 target, 4 agents) |
| XHack AIXHACK (SMC-PRIVATE) LIMITED · xhack.io | Commercial | Web apps, APIs, Networks, Active Directory, Cloud, Mobile, Code, LLM apps | SaaSDesktop | Individual AI plans $20 / $49 / $150 per month |
| AresAssail, Inc. · assailai.com | Commercial | Web apps, APIs, Mobile, LLM apps, AI agents | SaaSSelf-hosted | No published subscription price (demo/contact) |
| Astra Autonomous Pentest (Pentest Auto)Astra IT, Inc. (Astra Security) · getastra.com | Commercial | Web apps, APIs | SaaS | Pentest Auto $2,999/yr (1 target) |
| Beagle Securitybeaglesecurity.com | Commercial | Web apps, APIs | SaaS | Essential $119/month (2 tests/month), Advanced $359/month (15 tests/month) |
| Cobalt Autonomous PentestCobalt (Cobalt Labs, Inc.) · cobalt.io | Commercial | Web apps | SaaS | $3,500 per test, limited-time offer for web app tests started and completed by 31 Dec 2026 |
| CodeAnt AI PentestingCodeAnt AI, Inc. · codeant.ai | Commercial | Web apps, APIs, Code | SaaS | Not published |
| Escape AI PentestingEscape Technologies · escape.tech | Commercial | Web apps, APIs, Networks, LLM apps, AI agents, Code | SaaSSelf-hosted | On-Demand AI-Powered Pentest $3,000 (12-month contract) on Escape's own AWS Marketplace listing |
| FireCompassfirecompass.com | Commercial | Web apps, APIs, Mobile, Networks, Active Directory, LLM apps | SaaSSelf-hosted | Not published |
| HackerAIHackerAI LLC · github.com | Open core | — | SaaSDesktopCLI | 827 ★ · see licence |
| Helix CyberHelixML, Inc. · helix.ml | Commercial | Web apps, APIs, Networks, Code, Cloud, LLM apps | SaaS | Black box $3,000 |
| Invicti Agentic PentestInvicti Security (owner of Acunetix, now Invicti Web + API) · invicti.com | Commercial | Web apps, APIs | SaaS | $500 max per pentest (one application and its API suite) |
| NovaHadrian · hadrian.io | Commercial | Web apps | SaaS | From €1,250 per test excluding VAT |
| Ostorlab Agentic Deep Scan (Agentic Pentest)Ostorlab · ostorlab.co | Commercial | Mobile, Web apps, APIs, Code, Networks | SaaSSelf-hosted | Agentic Pentest one-time: Core $500, Advanced $2,000, Elite $4,000 (30-day retest window) |
| PenetrifyAlgofy s.r.o. · penetrify.cloud | Commercial | Web apps, APIs | SaaS | Starter $100/mo (1 test) |
| PensarPensarAI, Inc. · github.com | Open core | Web apps, APIs, AI agents, Mobile, Code | SaaSCLI | 323 ★ · Apache-2.0 |
| SableVulnetic Inc. · vulnetic.ai | Commercial | Web apps, APIs, Networks, Active Directory, Cloud, Mobile | SaaSSelf-hostedDocker | Practitioner $300/mo (web app, API, network and AD assessments, report, retests) |
| Sara PentestSynack · synack.com | Commercial | Web apps, Networks | SaaS | From $5,085 for 1 Standard Sara Pentest (one-time or recurring monthly) |
| Simbian AI Pentest AgentSimbian, Inc. · simbian.ai | Commercial | Web apps | SaaSSelf-hosted | Not published |
| SnipeStingrai Inc. · stingrai.io | Commercial | Web apps, APIs, LLM apps, Code | SaaS | Autonomous Pentest $3,000 per assessment (one web app + APIs) or US$650/month on a 12-month plan |
| StealthNet AIstealthnet.ai | Commercial | Web apps, APIs, Networks, LLM apps | SaaS | StealthStrike (AI-only) $1,500 for a single small scope |
| XBOWXBOW (XBOW USA Inc.) · xbow.com | Commercial | Web apps, APIs | SaaS | Quote-based, usage-based |
| Breach360BreachLock · breachlock.com | Commercial | Web apps, APIs, Networks | Self-hostedCLIDocker | Not published |
| Bright AI Pentesting (AI PT)Bright Security · brightsec.com | Commercial | Web apps, APIs | SaaS | Not published |
| NodeZeroHorizon3.ai · horizon3.ai | Commercial | Networks, Active Directory, Cloud, Web apps, APIs | SaaSSelf-hostedDocker | Per asset, 12-month term: Flex $15 per asset per test |
| NoveeNovee Security · novee.security | Commercial | Web apps, APIs, Mobile, LLM apps, AI agents | SaaSSelf-hosted | Not published |
| PenteraPentera (formerly Pcysys) · pentera.io | Commercial | Networks, Active Directory, Cloud, Web apps, APIs | No price on its own site | |
| RidgeBotRidge Security · ridgesecurity.ai | Commercial | Networks, Web apps, APIs, Active Directory | SaaSSelf-hosted | Not published |
| RidgeGenRidge Security · ridgesecurity.ai | Commercial | Web apps, APIs, Networks, Active Directory | Self-hosted | Not published |
| Snyk Evo Continuous Offensive SecuritySnyk · snyk.io | Commercial | Web apps, APIs, LLM apps, AI agents | SaaS | 4,000 credits per assessment (1 credit = $1) |
| SybilRunSybil · runsybil.com | Commercial | Web apps, APIs, Cloud, Code | SaaS | Not published |
| Terra PlatformTerra Security · terra.security | Commercial | Web apps, APIs, Networks, Cloud, Mobile, LLM apps, AI agents | SaaSSelf-hosted | Not published |
| Equixlyequixly.com | Commercial | Web apps, APIs, LLM apps | SaaS | Not published |
| Prancer (SwarmHack)Prancer · prancer.io | Commercial | Web apps, APIs, Networks, Active Directory, Cloud, Code, LLM apps, AI agents | SaaSSelf-hosted | Not published |
| AI red teaming · 35 · full entries | ||||
| Confident AIconfident-ai.com | Commercial | LLM apps, AI agents | SaaSSelf-hosted | Free $0 |
| Enkrypt AI Red TeamingEnkrypt AI · enkryptai.com | Commercial | Models, LLM apps, AI agents, MCP servers | SaaSSelf-hostedLibrary | Explore $0/month |
| Adversa AI Red TeamingAdversa AI Ltd. · adversa.ai | Commercial | Models, LLM apps, AI agents, MCP servers | SaaSSelf-hosted | Not published |
| ARTEMISRepello AI · repello.ai | Commercial | LLM apps, AI agents | SaaSLibrary | Not published |
| Cisco AI Defense (AI Validation)Cisco (from Robust Intelligence) · cisco.com | Commercial | Models, LLM apps, AI agents | SaaS | Not published |
| Giskard HubGiskard (Giskard AI SAS) · giskard.ai | Commercial | LLM apps, AI agents | SaaSSelf-hosted | Not published |
| Lakera Red (Check Point AI Red Teaming)Lakera (a Check Point Software Technologies company) · lakera.ai | Commercial | Models, LLM apps, AI agents | SaaS | Not published |
| Lasso Automated AI Red TeamingLasso Security · lasso.security | Commercial | LLM apps, AI agents, Models | SaaS | Not published |
| Mindgardmindgard.ai | Commercial | LLM apps, AI agents, Models, MCP servers | SaaSCLI | Not published |
| Promptfoo EnterprisePromptfoo (part of OpenAI) · promptfoo.dev | Commercial | LLM apps, AI agents, Models, MCP servers | SaaSSelf-hosted | Not published |
| Straiker Ascend AIStraiker · straiker.ai | Commercial | LLM apps, AI agents, Models, MCP servers | SaaSLibrary | Not published |
| TrojAI DetectTrojAI (A10 Networks) · a10networks.com | Commercial | Models, LLM apps, AI agents | Self-hosted | Not published |
| AI Red Teaming Agent (Microsoft Foundry)Microsoft · learn.microsoft.com | Commercial | Models, LLM apps, AI agents | SaaSLibrary | Not published |
| Dreadnode AI Red TeamingDreadnode · dreadnode.io | Commercial | Models, LLM apps, AI agents, MCP servers | SaaSSelf-hostedCLILibrary | Not published |
| F5 AI Red TeamF5 (from its CalypsoAI acquisition) · f5.com | Commercial | LLM apps, Models | SaaSSelf-hosted | Not published |
| HiddenLayer AI Attack SimulationHiddenLayer · hiddenlayer.com | Commercial | Models, LLM apps, AI agents | Not published | |
| Noma AI Red TeamingNoma Security · noma.security | Commercial | LLM apps, AI agents, MCP servers | SaaSCLI | Not published |
| Pillar Security RedGraph (Red Graph Suite)Pillar Security · pillar.security | Commercial | LLM apps, AI agents, MCP servers | SaaSSelf-hosted | Not published |
| Prisma AIRS AI Red TeamingPalo Alto Networks (technology from Protect AI) · paloaltonetworks.com | Commercial | Models, LLM apps, AI agents | SaaSSelf-hostedDocker | Not published |
| Prompt Security Automated AI Red TeamingSentinelOne (Prompt Security) · sentinelone.com | Commercial | LLM apps, AI agents | SaaSSelf-hosted | Not published |
| WonderBuild and WonderCheckAlice (formerly ActiveFence) · alice.io | Commercial | Models, LLM apps, AI agents | SaaS | Not published |
| Zscaler AI Red Teaming (formerly SPLX)Zscaler (SPLX, formerly SplxAI) · zscaler.com | Commercial | Models, LLM apps, AI agents, MCP servers | SaaSSelf-hosted | Not published |
| iFixAigithub.com | Open core | AI agents | CLILibrary | 44k ★ · Apache-2.0 |
| promptfooPromptfoo (part of OpenAI) · github.com | Open core | LLM apps, AI agents, Models, MCP servers, Code | CLILibrary | 27k ★ · MIT |
| garakNVIDIA · github.com | Open source | Models, LLM apps, AI agents | CLI | 9.9k ★ · Apache-2.0 |
| A.I.G (AI-Infra-Guard)Tencent Zhuque Lab · github.com | Open source | Models, AI agents, MCP servers, LLM apps | CLIDocker | 7.4k ★ · Apache-2.0 |
| Giskard (open source)Giskard · github.com | Open core | AI agents, LLM apps, Models | Library | 5.9k ★ · Apache-2.0 |
| PyRITMicrosoft (AI Red Team) · github.com | Open source | Models, LLM apps | CLILibrary | 4.8k ★ · MIT |
| DeepTeamConfident AI · github.com | Open core | LLM apps, AI agents, Models | CLILibrary | 3.1k ★ · Apache-2.0 |
| Agentic SecurityAgentic Security Team (msoedov) · github.com | Open source | LLM apps, AI agents, Models | CLI | 2.0k ★ · Apache-2.0 |
| FuzzyAICyberArk (a Palo Alto Networks company) · github.com | Open source | Models, LLM apps | CLI | 1.6k ★ · Apache-2.0 |
| promptmapUtku Sen (individual) · github.com | Open source | LLM apps, Models | CLI | 1.3k ★ · GPL-3.0 |
| Power PwnZenity (Michael Bargury) · github.com | Open source | AI agents, MCP servers | CLI | 1.3k ★ · MIT |
| AnamorpherTrail of Bits · github.com | Open source | LLM apps, AI agents, Models | Library | 1.1k ★ · Apache-2.0 |
| Agentic RadarSPLX (a Zscaler company) · github.com | Open source | AI agents, MCP servers | CLI | 1.1k ★ · Apache-2.0 |
| Open-source agent · 20 · full entries | ||||
| StrixStrix (OmniSecure, Inc.) · github.com | Open core | Web apps, APIs, Code | SaaSSelf-hostedCLIDocker | 74k ★ · Apache-2.0 |
| ShannonKeygraph (Keygraph Inc.) · github.com | Open core | Web apps, APIs, Code | SaaSSelf-hostedCLIDocker | 51k ★ · AGPL-3.0 |
| PentAGIVXControl · github.com | Open core | Web apps, Networks | SaaSSelf-hostedDocker | 34k ★ · MIT |
| PentestGPTGelei Deng (GreyDGL) and co-authors · github.com | Open source | CTFs, Web apps, Networks | CLIDocker | 16k ★ · MIT |
| HexStrike AI0x4m4 / OTT Cybersecurity LLC · github.com | Open source | Web apps, Networks, Cloud, CTFs | SaaSSelf-hostedMCP | 13k ★ · MIT |
| CAI (Cybersecurity AI)Alias Robotics · github.com | Open core | Web apps, APIs, Networks, CTFs | CLILibrary | 9.9k ★ · see licence |
| DecepticonBitter Security · github.com | Open core | Web apps, Networks, Active Directory, Cloud | SaaSSelf-hostedCLIDockerMCPLibrary | 6.6k ★ · Apache-2.0 |
| Cairnoritera (起零衍迹) · github.com | Open source | Networks, CTFs | Docker | 4.9k ★ · AGPL-3.0 |
| RedAmonSamuele Giampieri (samugit83) and contributors · github.com | Open source | Web apps, APIs, Networks, LLM apps | SaaSSelf-hostedDocker | 3.3k ★ · MIT |
| PentestAgentGH05TCREW · github.com | Open source | Web apps, Networks | CLIDocker | 3.3k ★ · MIT |
| Pentest Swarm AIArmur AI · github.com | Open source | Web apps, APIs, Networks, CTFs | CLIDocker | 3.2k ★ · AGPL-3.0 |
| Guardianzakirkun (Guardian Project) · github.com | Open source | Web apps, APIs, Networks, Active Directory, Cloud, Mobile, Code, LLM apps | CLI | 1.9k ★ · see licence |
| Pentest-AI0xSteph · github.com | Open source | Web apps | CLIMCP | 1.8k ★ · MIT |
| Pentest CopilotBugBase · github.com | Open core | Web apps, Networks, CTFs | SaaSSelf-hostedDocker | 1.6k ★ · MIT |
| NeuroSploitJoas A Santos · github.com | Open source | Web apps, Code, Networks, Active Directory, Cloud, LLM apps, Mobile | CLI | 1.5k ★ · MIT |
| hackingBuddyGPTTU Wien IPA-Lab (Andreas Happe, Jürgen Cito and contributors) · github.com | Open source | Networks, Web apps, APIs, Active Directory | CLILibrary | 1.3k ★ · MIT |
| DarkMoonASC-IT · github.com | Open core | Web apps, APIs, Networks, Active Directory, Cloud | SaaSSelf-hostedCLIDocker | 1.2k ★ · GPL-3.0 |
| NebulaBeryllium Security · github.com | Open source | Web apps, APIs, Networks | DesktopCLI | 1.1k ★ · BSD-2-Clause |
| VillagerCyberspike (Changchun Anshanyuan Technology Co., Ltd.) · pypi.org | Open source | Networks | Not published | |
| WhiteRabbitNeo (now DeepHat)Kindo (Usable Machines, Inc.) · deephat.ai | Open core | — | Model | Not published |
100 tools · 634 source URLs · checked November 8, 2026. Decima Systems, our editorial pick, is shown above and not in this table. Something wrong? Send a correction.
What each lane tests
How many tools in each lane say, on their own pages, that they test each kind of target. A tool can test several.
| Target | Platform /45 | AI red teaming /35 | Open-source agent /20 |
|---|---|---|---|
| Web apps | 44 | — | 17 |
| APIs | 37 | — | 9 |
| Networks | 21 | — | 16 |
| Active Directory | 9 | — | 5 |
| Cloud | 13 | — | 5 |
| Mobile | 12 | — | 2 |
| Code | 13 | 1 | 4 |
| LLM apps | 16 | 32 | 3 |
| AI agents | 9 | 31 | — |
| Models | — | 26 | — |
The penetration testing tools the agents drive
AI agents don't replace the classic penetration testing toolkit; they run it. 15 of the 20 open-source agents name the conventional tools they call. These are the ones named by two or more.
Counted from each agent’s own README. HexStrike AI claims “150+” tools and lists a selection; agents that say only “Kali tools” are not counted. Which agent drives what: open-source agents.
How to choose
Start from the target
Web apps and APIs, networks and Active Directory, or an LLM application need different tools. Filter by what you need tested before anything else.
Ask how findings are checked
Vendors range from “every engagement includes a final review by a human security expert” to “no human in the loop”. The entries quote each one.
Decide where it may run
SaaS sends traffic and findings through the vendor. Self-hosted platforms and open-source agents keep them with you, but you choose and pay for the model.
Read the licence
Open source is not always free for business use: some licences here forbid commercial use or require you to publish changes (AGPL).
Compare like with like
Prices come per test, per target, per seat, per asset or in credits. The pricing page lines up the one unit most vendors share.
Get written authorisation
Point a tool only at systems you own or have written permission to test, with the scope agreed in advance.
Questions
What is an AI pentesting tool?
Software that uses an AI model to plan or carry out parts of a penetration test: finding weaknesses, trying exploits and writing up what worked. This catalogue covers three kinds: commercial platforms that test web apps, networks or cloud accounts; tools that red-team LLM applications and AI agents; and open-source agents you run with your own model keys.
What are the best penetration testing tools?
This site doesn't rank tools on quality, because no outsider can measure it fairly. The classic toolkit (Nmap, Burp Suite, Metasploit, sqlmap, Nuclei) is still what the AI agents drive. Among open-source AI agents, the most-starred on GitHub on November 8, 2026 were Strix (74,375), Shannon (51,444), PentAGI (33,688).
Are there free AI pentesting tools?
Yes. All 20 open-source agents here are free to run, though you pay for the AI model they call, and 65 entries in total state a free tier, trial or free edition. Check the licence: some are free only for non-commercial use.
Can an AI tool replace a human penetration tester?
Not on the evidence published so far. In a study on a university network of about 8,000 hosts, the researchers’ ARTEMIS agent outperformed 9 of 10 professional testers, but AI agents produced more false positives and struggled with graphical interfaces. Most commercial platforms here say a person reviews findings, at least on some plans.
Is it legal to point these tools at a website?
Only at systems you own or have written permission to test. Running an attack tool against someone else’s system without authorisation is a crime in most countries, whether a person or an AI agent presses the button.
Sources for this page
- Each tool’s own site, repository and pricing page — Every fact in the table; 634 source URLs, linked from each entry on the lane pages.
- GitHub REST API — Stars, licences and release dates for the open-source entries.
- Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing (arXiv 2512.09882) — The agent-versus-professionals result in the questions.
- Decima Systems — The editorial pick’s facts and prices (we run it).
Each tool’s own sources are linked from its entry. How tools are chosen and ordered, and how to correct an entry: methodology.
Cite this page
AI Pentesting Tools. “AI pentesting tools, compared on what they publish.” Updated November 8, 2026. https://aipentestingtools.com/