Penetration testing tools are the scanners, proxies, password crackers and attack frameworks testers use to find weaknesses and prove they can be exploited: Nmap and Nessus for discovery, Burp Suite for web applications and APIs, BloodHound, Impacket and NetExec for Active Directory, Hashcat for passwords, and Metasploit or a command-and-control (C2) framework for exploitation. Most are free and open source. What you pay for in a penetration test is the senior person deciding where to point them and what the output means.
Below are the 30 pen testing tools that come up most on real engagements, grouped by phase of the penetration testing process: what each answers, what a hit means for you and how your logs would show it. Each category maps to one of our penetration testing services, run to our published methodology at fixed prices on our penetration testing pricing page.
How we chose these 30
Each tool is used on real engagements as of November 2026 (where one was archived, we list the successor: NetExec, not CrackMapExec). A tool made the list if it passed three tests:
- Current. It is in use on engagements now, and it either still receives updates or still works as last released, as Kerbrute and Mimikatz do.
- A distinct question. We list one or two tools per job, not every alternative, which is why overlapping tools such as Hydra and Nikto are left out.
- Something a defender can act on. Its use leaves evidence you can log, or its findings map to a specific fix.
NIST’s SP 800-115 separates target identification and analysis (network discovery, port and service identification, vulnerability scanning), generally automated, from target vulnerability validation (password cracking, penetration testing, social engineering), which corroborates that a vulnerability exists, manually or with tools depending on the technique and the test team’s skill.
Network penetration testing tools: discovery and scanning
| # | Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|---|
| 1 | Nmap | Free, open source | Which hosts, ports, services and versions answer | Services nobody meant to expose, or versions with known flaws |
| 2 | Masscan | Free, open source | What is open across huge address ranges, fast | Forgotten hosts in space you assumed was empty |
| 3 | OWASP Amass | Free, open source | Which domains, subdomains and ranges belong to you | An attack surface larger than your asset register |
| 4 | Wireshark | Free, open source | What is actually crossing the wire | Cleartext credentials and legacy protocols |
| 5 | Nessus | Commercial; 30-day non-commercial Essentials license | Which known CVEs and missing patches each host carries | Patch debt, before anyone proves it is exploitable |
| 6 | Nuclei | Free, open source | Whether a target matches thousands of community detection templates | A known exposure, often a newly published one |
Our network penetration testing guide shows how these fit a full engagement, from scoping to report.
Shopping for free scanning software? As of November 2026 Tenable describes Nessus Essentials as a 30-day license for up to 5 IPs, strictly non-commercial, so a business needs Nessus Professional or the free, open-source Greenbone Community Edition (OpenVAS). Our Invadel vs Nessus page covers what a scanner proves, and our vulnerability scanning service runs one with an analyst validating every result, from $1,500 for up to 250 devices.
Web penetration testing tools
| # | Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|---|
| 7 | Burp Suite | Community free; Professional paid | Every request between browser and server, readable and modifiable | Whatever the tester proves: broken access control, injection, logic flaws |
| 8 | ZAP | Free, open source | The same proxy model, plus scans that fit a CI pipeline | Regressions caught between manual tests |
| 9 | ffuf | Free, open source | Unlinked content, and whether IDs and forms survive thousands of attempts | Backup files, staging hosts, old API versions, missing rate limits |
| 10 | Gobuster | Free, open source | Hidden directories, subdomains and virtual hosts | Content you thought was private because nothing linked to it |
| 11 | sqlmap | Free, open source | How far a suspected SQL injection reaches | The database behind the parameter, sometimes the server |
Burp Suite from PortSwigger is the standard. The free Community Edition lacks the scanner, the full Intruder and saved projects, so professionals buy Professional; we use it on nearly every web engagement, as our Burp Suite comparison explains. ZAP, now branded ZAP by Checkmarx and still open source, is a common pick for pipeline scanning. See also our guides to DAST tools and web application security testing.
Active Directory and internal network tools
| # | Tool | What it does | What it means when it works |
|---|---|---|---|
| 12 | Responder | Answers LLMNR, NBT-NS and mDNS lookups to capture credential hashes | Legacy name resolution is still on |
| 13 | Kerbrute | Confirms usernames and tests passwords through Kerberos | Usernames can be enumerated and weak passwords sprayed, unseen unless someone watches 4768 and 4771 |
| 14 | NetExec | Tests where a credential works across hosts and protocols | Credentials are reused or over-privileged |
| 15 | Impacket | Relay, Kerberoasting, remote execution and credential dumping over Windows protocols | SMB signing is off or admin rights are too broad |
| 16 | BloodHound | Graphs Active Directory and finds the shortest path to Domain Admin | Ordinary permissions chain into full control |
| 17 | Rubeus | Requests and abuses Kerberos tickets | Kerberoastable accounts or delegation flaws |
| 18 | Certipy | Finds misconfigured AD Certificate Services templates | A low-privilege user can get a certificate that logs in as someone else |
| 19 | Mimikatz | Extracts credentials from Windows memory | Credentials sit where an admin-level attacker can read them |
| 20 | Evil-WinRM | Opens a shell over Windows Remote Management | A stolen credential carries remote-management rights |
All nine have free, open-source editions, and most of the time they exploit no missing patch: they use Windows as designed, with credentials or permissions they should not have. Most serve one goal: privilege escalation from an ordinary account to one that controls the domain. The fixes are structural (disable LLMNR and NBT-NS, enforce SMB signing, retire RC4, tier admin accounts, lock down certificate templates); our Active Directory hardening checklist and explainers on Kerberoasting, NTLM relay and PetitPotam, golden and silver tickets and lateral movement go deeper.
Attackers use them too. As of November 2026, MITRE ATT&CK lists 52 threat groups using Mimikatz and 20 using Impacket, including Volt Typhoon and Scattered Spider, and CISA’s LockBit advisory (AA23-165A) names BloodHound, Impacket and Mimikatz among the tools LockBit affiliates used.
Password cracking tools
| # | Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|---|
| 21 | Hashcat | Free, open source | Which captured hashes crack offline on GPUs | Passwords that meet policy and still fall to a wordlist |
| 22 | John the Ripper | Free, open source; Pro edition paid | The same, with wide support for protected files and keys | Weak passwords on documents, archives and private keys |
Cracking happens offline, so your environment never sees it; what you can see is the hash being captured or the ticket requested. A good report lists the password patterns that fell, not the passwords.
Exploitation and command-and-control frameworks
| # | Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|---|
| 23 | Metasploit Framework | Free, open source (Rapid7 and community); Pro edition paid | Whether a known vulnerability becomes code execution | A missing patch is exploitable, not theoretical |
| 24 | Cobalt Strike | Commercial (Fortra) | Whether an implant can operate for weeks unnoticed | Endpoint detection, egress filtering and the SOC missed sustained activity |
| 25 | Sliver | Free, open source (Bishop Fox) | The same, with an open-source C2 | As above |
Metasploit’s payload generator has its own msfvenom explainer. Cobalt Strike is sold for adversary simulation and widely abused: ATT&CK lists 30 groups using Cobalt Strike, and in 2023 Microsoft’s Digital Crimes Unit, Fortra and Health-ISAC obtained a court order against cracked legacy copies that Microsoft linked to more than 68 ransomware attacks on healthcare organizations in more than 19 countries. ATT&CK lists APT29 among the groups using Sliver. Our explainers on C2 servers and red team tools cover the rest of the kit.
Phishing, wireless, cloud and mobile tools
| # | Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|---|
| 26 | GoPhish | Free, open source | Who clicks, submits credentials or reports, by department | Reporting habits, measured instead of assumed |
| 27 | Evilginx | Free, open source; Pro edition paid | Whether an adversary-in-the-middle page steals a session past MFA | Your MFA is phishable; passkeys or FIDO2 fix it |
| 28 | Aircrack-ng | Free, open source | Whether a Wi-Fi pre-shared key can be recovered | A weak key that puts an outsider on the network |
| 29 | Prowler | Free, open source; paid cloud edition | Which settings fail checks across AWS, Azure, Google Cloud, Kubernetes and Microsoft 365 | Misconfigurations, before anyone shows which chain together |
| 30 | Frida | Free, open source | What an app does at runtime with its client-side protections off | Secrets and logic the app trusted the device to protect |
Frida, usually paired with MobSF, is how a tester strips certificate pinning and root detection to reach the API behind an app, which is why those controls never replace server-side checks. Our Android penetration testing guide walks through that work.
API and cloud penetration testing tools
Four more tools sit outside the 30 but come up whenever the scope is an API or a cloud account:
| Tool | Free or paid | What it answers | What a hit means for you |
|---|---|---|---|
| Postman | Free plan; paid tiers | What each documented endpoint returns when requests from an imported API definition are replayed as a different user | Endpoints that trust the client to ask only for its own records |
| Kiterunner | Free, open source (Assetnote) | Which API routes and endpoints exist that the documentation never mentions | Old versions and internal endpoints reachable from the internet |
| Pacu | Free, open source (Rhino Security Labs) | How far one set of AWS credentials can be pushed | Over-broad IAM permissions, shown as a working path |
| Scout Suite | Free, open source (NCC Group) | Which settings look risky across AWS, Azure and Google Cloud | The same class of misconfiguration Prowler reports, from a second engine |
Route discovery looks like web fuzzing in your logs: bursts of 404s on API paths from one client. Pacu works through the AWS API with the credentials it is given, so AWS CloudTrail records what it enumerates and changes. Our AWS penetration testing guide covers the cloud side in depth.
How to detect penetration testing tools in your environment
Every tool above leaves traces. Wire up these signals before a test, the cheapest time to learn which alerts fire.
| Activity (tools) | Log source | Event or signal | What it indicates |
|---|---|---|---|
| Port scanning (Nmap, Masscan) | Firewall, NetFlow, IDS | One source touching many ports or hosts within minutes | Discovery underway |
| Web fuzzing (ffuf, Gobuster, sqlmap) | Web server, CDN and WAF logs | Bursts of 404s or database errors from one client; default tool User-Agents | Content discovery or injection testing |
| Poisoning (Responder) | Network monitoring | LLMNR or NBT-NS replies from a host that is not a name server | Credential capture on the segment |
| Spraying (Kerbrute) | Domain controller Security log | 4768 failures with code 0x6; 4771 with code 0x18 across many accounts | Username enumeration and password spraying |
| Kerberoasting (Rubeus, Impacket) | Domain controller Security log | Bursts of 4769 for many services from one account; type 0x17 (RC4) tickets, which since Microsoft’s 2026 RC4 changes go only to accounts still set to allow RC4 | Service-account hashes requested for cracking |
| Lateral movement (NetExec, Impacket, Evil-WinRM) | Endpoint Security, System and PowerShell logs | 4624 logon type 3 across many hosts; 7045 service installs; 4698 scheduled tasks; 4104 script blocks | A credential used to execute remotely |
| DCSync (Mimikatz, Impacket) | Domain controller Security log | 4662 showing replication rights used by an account that is not a domain controller | The password database being copied |
| Memory credential theft (Mimikatz) | Sysmon, EDR | Sysmon Event ID 10 targeting lsass.exe | An attempt to read credentials from LSASS |
| Certificate abuse (Certipy) | CA Security log and database; domain controller Security log | 4886 and 4887 (request received, certificate issued), written only when the CA’s own audit filter is on; an alternative name may not show in the event text, so check issued names against requesters in the CA database; 4768 certificate logons by privileged accounts | A certificate that impersonates another user |
| C2 traffic (Cobalt Strike, Sliver) | Proxy, DNS, Sysmon | Regular-interval callbacks to young domains; Sysmon 3, 17, 18 and 22 | An implant checking in |
| Session theft (Evilginx) | Identity provider sign-in logs | One session used from a new IP or country minutes after sign-in | A stolen session cookie replayed |
Microsoft’s Sysmon documentation says Event ID 10 enables detection of tools that read LSASS memory to steal credentials (filter it, or it gets noisy). The domain controller events need advanced audit policies switched on. On RC4, Microsoft’s CVE-2026-20833 guidance sets out the change: with updates released from April 2026 the domain controller defaults to AES for accounts with no explicit encryption setting, and honors RC4 only where an administrator has explicitly configured it. On the certificate authority, 4886 and 4887 are written only when both the Audit Certification Services subcategory and the CA’s own audit filter (the Auditing tab of its properties) are enabled, as our Certipy guide explains. A test that raises no alerts is a finding in its own right, and our purple team assessment runs these techniques with your defenders watching.
Open source and free penetration testing software you can run yourself
| Tool | Run it in-house? | What it will not tell you |
|---|---|---|
| Nmap | Yes, on ranges you own | Whether an exposed service is exploitable |
| A vulnerability scanner (OpenVAS, Nessus Professional) | Yes, on a schedule | Chains, logic flaws, false positives |
| ZAP | Yes, against staging in CI | Authorization and business logic |
| Prowler | Yes, read-only | Which misconfigurations chain into account takeover |
| BloodHound | Yes, with security team sign-off | Proof that a path works |
| Responder, Mimikatz, C2 frameworks | Only inside an authorized engagement | They trip EDR, can disrupt production and need written rules of engagement |
Run tools only against systems you own or have written permission to test: unauthorized access is a crime under the federal Computer Fraud and Abuse Act (18 U.S.C. 1030) and state law. And a scan you run on yourself is not the independent evidence auditors and enterprise customers usually ask for.
Penetration testing software vs a penetration test
Pentest tools are good at breadth: known CVEs, open ports, default paths, failed benchmarks. They are poor at findings that need context, often the most serious ones:
- Authorization. A tool cannot tell whether user A should see user B’s invoice; an IDOR vulnerability takes two accounts and a person who understands the product.
- Business logic. Skipping a payment step, reusing a single-use coupon, approving your own request.
- Chains. Three medium findings that together reach Domain Admin. BloodHound draws the path; a tester proves it and stops where the rules of engagement say.
Our guides to manual vs automated penetration testing and penetration testing vs vulnerability scanning go deeper, and we compare autonomous penetration testing vendors separately. Online “automated pentest” platforms mostly sit closer to the scanner. One buyer check: if a “penetration test” report is mostly scanner output under a cover page, you bought a scan.
How Invadel uses these tools, and what it costs
What you buy from us is a senior in-house tester, never a subcontractor, driving these tools inside a written scope, and every finding verified by hand with a CVSS rating, reproduction steps and the fix. Source addresses and testing windows are agreed in advance, so your team can match its alerts to our timeline.
| Engagement | Tools from this list that fit the scope | From |
|---|---|---|
| External network | Nmap, Masscan, Amass, Nuclei, a vulnerability scanner | $4,200 |
| Web application | Burp Suite Professional, ffuf, Gobuster, sqlmap | $5,200 |
| API | Burp Suite Professional, ffuf | $4,000 |
| Internal network | Responder, Kerbrute, NetExec, Impacket, BloodHound, Evil-WinRM | $6,000 |
| Active Directory assessment | BloodHound, Rubeus, Certipy, Impacket, Hashcat | $6,000 |
| Cloud | Prowler-style configuration checks plus manual attack paths | $6,800 |
| Mobile application | Frida, MobSF, Burp Suite Professional | $6,000 |
| Wireless | Aircrack-ng suite | $3,800 |
| Phishing | Campaign tooling, adversary-in-the-middle pages | $3,600 (no retest) |
| Red team | A C2 framework, phishing, and the external and internal network tools above | $12,500 |
Each price is a published starting point, fixed in writing from the details you send through our scoping questionnaire, with no sales call. Every penetration test includes executive and technical reports, an attestation letter, the findings platform and a free retest of remediated findings, and onboarding begins within 24 hours of a signed proposal. The pricing page shows the size tiers, the cost calculator adds tests together, and the sample report shows a finished finding.
Frequently asked questions
What are the top 10 tools for pentesting? For most engagements: Nmap, Burp Suite, a vulnerability scanner such as Nessus, Metasploit, BloodHound, Impacket, NetExec, Responder, Hashcat, and ffuf or Gobuster. Which ones matter depends on the scope.
Which penetration testing tools are free? Almost all of them. Of the 30 here, only Cobalt Strike has no free edition. The tightest limits are on Burp Suite Community (no scanner, a throttled Intruder) and Nessus Essentials (30 days, 5 IPs, non-commercial); Metasploit, Prowler, Evilginx and John the Ripper keep extra features for paid editions.
Which penetration testing software do you have to pay for? Of the 30, Cobalt Strike and Nessus are commercial products, and Burp Suite Professional is the paid edition most web testers need. Metasploit Pro, John the Ripper Pro, Evilginx Pro and Prowler’s cloud edition are optional paid tiers. Everything else is free and open source.
What is the difference between a vulnerability scanner and a penetration testing tool? A scanner such as Nessus or OpenVAS checks hosts against a database of known flaws and reports what might be exploitable. Tools such as Burp Suite, Impacket and Metasploit are driven by a person to prove what is exploitable. A penetration test uses both.
Is it legal to use penetration testing tools? In the US, owning them is legal; using them against systems you do not own or have written authorization to test is not. Professional tests run under a signed scope and rules of engagement.
Do penetration testers use Kali Linux? Often, for convenience: Kali and Parrot OS bundle most of the free tools here. The operating system matters far less than the scope, the tester’s judgment and the evidence.
Can a tool replace a penetration test? No. Tools find known patterns at scale but miss broken authorization, business logic and chained attack paths.
The short version
Penetration testing tools are mostly free, and attackers have the same ones. What separates a real test from a tool run is the person steering them, the chaining and the evidence. For that at a fixed price with a free retest, see our penetration testing services and published prices, or scope a test and get a written number the same day.
Written by
Mark Kiss, Founder & CEO
15+ years in the field. Focus: Offensive security & strategy. Part of the team that scopes, tests, and reports for our clients. Full profile →