DETECT hunt Library

Library health

What each source ships with, as a share of its detections. 31,964 of 62,921 detections carry no ATT&CK technique, so the coverage map leaves them out. 10,088 of them are YARA rules, which match malware families in files and memory and name no technique. The 1,208 that name a tactic still appear on that tactic's page, under No technique tagged.

Framework v0.136.0, updated November 17, 2026.

SourceDetectionsATT&CK techniqueDescriptionSeverityQuery textKeys-on valuesTranslator reads itRobustness draftFired on a recording
Emerging Threats Open10,313100%100%100%100%48%0%0%-
ProjectDiscovery search queries7,7770%100%0%100%1%0%0%-
signature-base YARA5,8520%99%0%100%41%0%0%-
Microsoft Sentinel4,98478%100%59%100%22%75%74%-
Snort 3 community rules4,0244%100%100%100%42%0%0%-
Wazuh ruleset3,84027%100%100%100%3%0%0%-
SigmaHQ3,76687%100%100%100%67%100%100%9%
Elastic protections YARA3,0690%100%0%100%11%0%0%-
Elastic detection rules2,21696%100%100%95%41%66%66%-
Splunk Security Content2,09099%100%0%100%53%59%59%-
Anvilogic Forge armory1,898100%100%0%100%57%71%71%-
Exabeam content library1,424100%100%0%0%1%0%0%-
Elastic protections1,33299%100%0%100%63%30%30%-
Sublime Security1,2910%100%100%100%1%0%0%-
Panther analysis86669%97%100%100%17%24%23%-
Malicious infrastructure hunting queries8640%100%0%100%3%0%0%-
Agent Threat Rules84427%100%100%100%39%0%0%-
LOLRMM716100%100%100%100%58%100%99%0%
reprise99 Sentinel queries4620%100%0%100%30%66%65%-
Google SecOps (Chronicle)37978%100%100%100%36%0%0%-
ep3p Sentinel KQL3540%28%0%100%27%53%53%-
mdecrevoisier Sigma rules33799%100%100%100%98%100%71%5%
ReversingLabs YARA3090%100%0%100%0%0%0%-
Bert-JanP hunting queries30233%91%0%100%37%57%57%-
Microsoft Defender hunting2662%98%0%100%58%77%77%-
JPCERT/CC YARA2390%100%0%100%16%0%0%-
Micah Babinski Sigma rules22394%100%100%100%80%100%100%-
Alex Verboon hunting queries19823%94%0%100%27%73%72%-
osquery packs1960%100%0%100%7%0%0%-
Hayabusa rules19120%85%100%100%99%100%50%15%
Mandiant red team tool countermeasures1720%62%0%100%8%0%0%-
AIPT detection pack15533%100%88%100%10%35%35%-
Volexity threat intel (YARA)1490%100%0%100%18%0%0%-
Elastic hunting queries14397%100%0%100%33%5%5%-
bartblaze YARA1420%100%0%100%19%0%0%-
CrowdStrike LogScale community13624%30%0%100%14%30%30%-
AI DFIR toolkit13624%100%44%100%26%29%28%0%
Falco plugin rules13412%100%100%100%9%0%0%-
ESET malware IOCs (YARA)1170%86%0%100%20%0%0%-
Joe Security Sigma rules11212%100%100%100%41%100%99%11%
Google Cloud security analytics10297%100%100%100%6%0%0%-
MITRE CAR9999%100%0%100%79%62%62%-
Falco rules9585%100%100%100%8%0%0%-
cyb3rmik3 KQL hunting queries8738%100%0%100%45%68%67%-
Awesome Censys queries840%100%0%100%0%0%0%-
Jamf Protect custom analytics801%100%100%100%0%0%0%-
GitLab threat detections (TLDR)6179%100%100%100%20%100%38%-
Awesome Shodan queries540%100%0%100%0%0%0%-
benscha KQL advanced hunting51100%100%0%100%18%84%80%-
FalconForce FalconFriday48100%100%0%100%83%62%62%-
Okta customer detections34100%100%0%100%76%0%0%-
Tracee detectors3397%100%100%100%0%0%0%-
OTRF ThreatHunter-Playbook2669%100%0%100%100%0%0%-
Auth0 customer detections26100%100%88%100%0%100%96%-
CTID cloud analytics16100%100%100%100%6%100%56%-
vector-sec Sublime rules70%100%86%100%0%0%0%-
All sources62,92149%99%52%98%32%24%24%8%

Columns

ATT&CK technique
Detections tagged with a technique the pinned ATT&CK version knows, counting a revoked technique as the one that replaced it.
Severity
The share of detections with a severity level the catalog can map. It reads 0% where a source rates its rules on another scale. Splunk scores risk per rule, and the YARA sources carry a 0 to 100 score that the library shows as the score draft instead. Elastic protections, osquery, and MITRE CAR carry no rating.
Keys-on values
Detections with at least one program, event ID, registry key, action, or CVE read from the query (observables).
Translator reads it
Sigma rules and queries the translator can parse into at least one other language.
Robustness draft
Detections with a first-pass level of 1 to 3, scored the way Summiting the Pyramid scores an analytic: each observable the rule matches gets a level, and the condition combines them.
Fired on a recording
Of the Windows Sigma rules that have a recording of their technique in OTRF Security-Datasets, the share that fired on it. The denominator is those rules, 1,689 in all, so sources with no Windows Sigma rules show a dash.

Checks against outside sources

The robustness draft was checked against the 72 SigmaHQ rules MITRE's CTID scored by hand: it matched the hand score on 55 and stayed at or under it on 66. The Windows Sigma rules were run against 107 recordings of attacks, 776,654 events over 56 techniques: of the 1,695 rules with a recording of their technique, 139 fired on it. The outside checks page has the full numbers and what they do and do not show.

Every figure is computed from the catalog on each build. This build read the catalog on 2026-10-05.