↓ Skip to main content
Security

MITRE ATT&CK, plainly: tactics, technique IDs and a cheat card for each tactic

A football coach studies match footage of the other team: how they open, how they press, how they score. MITRE ATT&CK is the match footage of real attackers, cut into named and numbered moves.

The OWASP Top 10 describes weaknesses in your software. ATT&CK describes what attackers do: how they get in, stay in, move around and take what they came for. You will see its IDs, such as T1190 or T1552.001, in threat reports, detection rules and, increasingly, in pipeline designs like the one in my SSDLC post. This post explains how to read them.

What it is
#

MITRE’s own definition is short: ATT&CK is a “knowledge base of adversarial techniques based on real-world observations” (ATT&CK resources). It is free to use, and every entry links to the public reports it was built from.

There are three matrices: Enterprise (Windows, Linux, macOS, cloud, containers, network devices), Mobile and ICS (industrial control systems). This post covers Enterprise.

The current release is ATT&CK v20, from 13 October 2026. Its Enterprise matrix has 15 tactics, 227 techniques and 479 sub-techniques (ATT&CK updates).

How to read an ID
#

ATT&CK has a small vocabulary. MITRE puts it this way: tactics are the why, techniques are the how.

PrefixIs a…ExampleAnswers
TATactic: the attacker’s goal at this stepTA0001 Initial AccessWhy are they doing this?
TTechnique: a way to reach that goalT1552 Unsecured CredentialsHow, in general?
T….nnnSub-technique: a more specific wayT1552.001 Credentials In FilesHow, exactly?
MMitigation: a defence that helpsM1047 AuditWhat stops it?
G, SGroup and Software: who uses it, with which toolsWho has done it?

So T1552.001 reads as: technique 1552, Unsecured Credentials, sub-technique 001, Credentials In Files (T1552.001). It falls under the Credential Access tactic, and MITRE lists mitigations such as auditing for password files (M1047) and password policies that forbid storing them (M1027). That is a leaked .env file in a repository, described in a way every security tool understands.

One technique can serve several tactics. T1078 Valid Accounts, logging in with stolen but real credentials, is listed under Initial Access, Persistence, Privilege Escalation and Stealth, because a real account helps with all four.

What changed recently: Defense Evasion was split
#

If you learned ATT&CK a few years ago, you will remember 14 tactics, including Defense Evasion. That tactic is gone. In its place are two (Sophos):

  • Stealth (TA0005, the old ID): “The adversary is trying to hide and conceal their actions, appearing as normal behavior”, for example Masquerading (T1036) or Obfuscated Files or Information (T1027) (TA0005).
  • Defense Impairment (TA0112, new, created in April 2026): “The adversary is trying to break security mechanisms, pipelines, and tooling so defenders can’t see or trust what’s happening”, for example Disable or Modify Tools (T1685) (TA0112).

The difference is useful: hiding from your defences versus breaking them. Note the word “pipelines” in the new tactic. Tampering with a CI system’s security checks is now a named attacker goal. Older reports and tools still use Defense Evasion; map them to one of the two.

The cheat cards: 15 tactics
#

One card per tactic, in the order an attack usually unfolds. Each has the tactic ID, its goal in plain words, one example technique and what a development team can do about it. The colours group the phases: blue for getting in, orange for gaining a foothold, purple for spreading, red for the payoff.

TA0043 1 · getting in

Reconnaissance

Goal: learn about the target.

Example: T1595 Active Scanning of your public hosts.

Dev angle: expose less; no version banners or debug endpoints.

TA0042 2 · getting in

Resource Development

Goal: prepare tools and infrastructure.

Example: T1583 Acquire Infrastructure, such as look-alike domains.

Dev angle: mostly outside your control; watch for typo-squatted packages.

TA0001 3 · getting in

Initial Access

Goal: get a first foothold.

Example: T1190 Exploit Public-Facing Application; T1195 Supply Chain Compromise.

Dev angle: patch exposed services first; scan dependencies.

TA0002 4 · foothold

Execution

Goal: run their code.

Example: T1059 Command and Scripting Interpreter.

Dev angle: never pass user input to a shell; parameterise.

TA0003 5 · foothold

Persistence

Goal: stay in after a restart or a password change.

Example: T1053 Scheduled Task/Job.

Dev angle: immutable images; rebuild rather than patch in place.

TA0004 6 · foothold

Privilege Escalation

Goal: get more rights than they started with.

Example: T1068 Exploitation for Privilege Escalation.

Dev angle: run as non-root; least-privilege service accounts.

TA0005 7 · foothold

Stealth

Goal: hide, and look like normal activity.

Example: T1036 Masquerading as a legitimate binary.

Dev angle: signed artifacts; know what normal looks like.

TA0112 8 · foothold · new

Defense Impairment

Goal: break the defences, including pipelines.

Example: T1685 Disable or Modify Tools.

Dev angle: protect CI config; alert when a security check is removed.

TA0006 9 · spreading

Credential Access

Goal: steal account names and passwords.

Example: T1552.001 Credentials In Files.

Dev angle: secret scanning on every commit; a secrets manager.

TA0007 10 · spreading

Discovery

Goal: learn the environment from inside.

Example: T1082 System Information Discovery.

Dev angle: minimal images with no extra tools to explore with.

TA0008 11 · spreading

Lateral Movement

Goal: move to other systems.

Example: T1021 Remote Services.

Dev angle: network policies; no shared credentials between services.

TA0009 12 · payoff

Collection

Goal: gather the data they want.

Example: T1213 Data from Information Repositories, such as wikis and code hosts.

Dev angle: access reviews; nothing secret in tickets or wikis.

TA0011 13 · payoff

Command and Control

Goal: talk to compromised systems.

Example: T1071 Application Layer Protocol, hiding in normal web traffic.

Dev angle: restrict outbound traffic from workloads.

TA0010 14 · payoff

Exfiltration

Goal: get the data out.

Example: T1041 Exfiltration Over C2 Channel.

Dev angle: egress limits; alert on unusual data volumes.

TA0040 15 · payoff

Impact

Goal: disrupt, destroy or hold to ransom.

Example: T1486 Data Encrypted for Impact (ransomware).

Dev angle: tested backups; infrastructure you can rebuild from code.

How a developer uses it
#

ATT&CK was built for defenders and threat hunters, but it is useful one step earlier, in the pipeline.

  1. Name what each control stops. “Gitleaks on commit” is a tool. “Gitleaks on commit stops T1552.001” is a reason. The SSDLC post tags every pipeline level this way:

    Pipeline levelTechnique it guards againstMITRE mitigation
    Commit: secret scanningT1552.001 Credentials In FilesM1047 Audit
    PR and build: dependency scanningT1195.001 Compromise Software Dependencies and Development ToolsM1016 Vulnerability Scanning
    Build: signed provenanceT1195.002 Compromise Software Supply ChainM1013 Application Developer Guidance
    Nightly rescan of exposed appsT1190 Exploit Public-Facing ApplicationM1051 Update Software

    Techniques from T1195 and T1190; mitigations as MITRE lists them.

  2. Find the gaps. Shade the techniques you cover in the free ATT&CK Navigator. The white areas are the conversation for your next planning meeting.

  3. Speak the same language as the security team. A finding tagged with an OWASP category (the weakness) and an ATT&CK technique (the attack) is understood by developers and by the people who write detections.

What it does not do
#

  • It does not rank anything. ATT&CK does not say which technique is most dangerous for you. Use exposure, KEV and EPSS for that, as in part 4 of the standards series.
  • It is not a checklist to complete. Hundreds of techniques, many irrelevant to a given system. Pick the ones your architecture makes possible.
  • IDs change. Techniques get added, moved and occasionally replaced; in 2026 even a tactic was split. Note the version when you tag findings.

Takeaways
#

  • ATT&CK is match footage of real attackers: tactics are the why (TA…), techniques the how (T…), sub-techniques the exact how (T….001).
  • v20 (13 October 2026): 15 Enterprise tactics, 227 techniques, 479 sub-techniques. Defense Evasion became Stealth and Defense Impairment.
  • For developers, the value is naming what each control stops and spotting what nothing stops.
  • OWASP Top 10 describes the weakness; ATT&CK describes the attack. Use both.

References
#

Author
Samarthya
Engineer, learner, tinkerer. This is my trophy shelf of everyday work and leisure.

Related

OWASP Top 10:2025, plainly: the ten risks and a cheat card for each

A pilot’s checklist does not list every way a plane can fail. It lists the few that keep happening, in the order that matters. The OWASP Top 10 is that checklist for web applications. When a scanner says a finding is “A05” or a security review mentions “Top 10 coverage”, this is the list they mean. This post is a quick introduction: what the list is, what changed in 2025, why each category matters, and a cheat card for each one.