threat-hunting
Use when hunting threats or engineering detections — ATT&CK Detection-Strategies, Sigma + correlation with Detection-as-Code CI, Windows endpoint hunting (Sysmon/ETW/LSASS/LOLBins), network C2 hunting (JA4+, beaconing, DNS tunneling), cloud-identity hunting, Atomic Red Team purple-team validation
Install / Use
npx skills add hypnguyen1209/offensive-claude --skill threat-huntingInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of threat-hunting
threat-hunting scores 86/100 on our quality scale, 2308th of 4,619 Development & Engineering skills we index (top 50%).
Its SKILL.md is 9.2 KB long, well organised into 13 sections with 1 code example: a thorough specification that gives an agent plenty to work with.
It has 377 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated 14 days ago, so threat-hunting is actively maintained.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.
Automated pattern scan on 2026-10-05. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
threat-hunting compared with similar skills
All 4 of these similar skills score higher than threat-hunting; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| threat-hunting (this skill)by hypnguyen1209 | 86 | 377 | 14d ago | SKILL.md |
| ai-job-searchby MadsLorentzen | 100 | 45.0k | 1d ago | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 41.7k | 4d ago | CLAUDE.md |
| algorithmic-artby anthropics | 100 | 177.9k | 12d ago | SKILL.md |
| pptxby anthropics | 100 | 177.9k | 12d ago | SKILL.md |
Frequently asked questions
- How do I install threat-hunting?
- Run
npx skills add hypnguyen1209/offensive-claude --skill threat-hunting. The install tabs above show the steps for each supported agent. - Which AI agents does threat-hunting work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is threat-hunting safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is threat-hunting still maintained?
- The repository was last updated 14 days ago, so threat-hunting is actively maintained.
Skill content
View source on GitHubname: threat-hunting description: Use when hunting threats or engineering detections — ATT&CK Detection-Strategies, Sigma + correlation with Detection-as-Code CI, Windows endpoint hunting (Sysmon/ETW/LSASS/LOLBins), network C2 hunting (JA4+, beaconing, DNS tunneling), cloud-identity hunting, Atomic Red Team purple-team validation metadata: type: defensive phase: detection tools: sigma-cli, pysigma, hayabusa, chainsaw, velociraptor, sysmon, zeek, rita, ja4, atomic-red-team, caldera, splunk, sentinel, defender, kql mitre: TA0043 kill_chain: phase: [report] step: [8] attck_tactics: [TA0043, TA0042, TA0011, TA0006, TA0005] attck_techniques: [T1059.001, T1003.001, T1562.001, T1562.002, T1218, T1105, T1071, T1071.001, T1071.004, T1571, T1572, T1528, T1550.001, T1078.004, T1098, T1543.003, T1070.001, T1558.003, T1003.006] depends_on: [red-team-ops, incident-response] feeds_into: [] inputs: [finding_records, log_data, ioc_list, evtx, zeek_logs, cloudtrail, sigin_logs] outputs: [sigma_rules, correlation_rules, attck_navigator_export, coverage_matrix, detection_report, hunt_findings] references:
- references/methodology-hunt-loop.md
- references/windows-endpoint-hunting.md
- references/sigma-rule-engineering.md
- references/network-c2-hunting.md
- references/cloud-identity-hunting.md
- references/purple-team-validation.md scripts:
- scripts/dac_validate.py
- scripts/evtx_hunt.py
- scripts/beacon_hunter.py
- scripts/cloudtrail_hunt.py
- scripts/coverage_matrix.py
- scripts/entra_hunt.kql
- scripts/sigma_pipeline.sh
- scripts/sysmon_config_2025.xml
Threat Hunting & Detection Engineering
When to Activate
- Hypothesis-driven hunting across endpoint, network, cloud, and identity telemetry
- Writing & shipping detections (Sigma + correlation) as version-controlled code in CI
- Mapping & measuring coverage against MITRE ATT&CK v18 (Detection Strategies / Analytics)
- Hunting Windows post-exploitation: ETW/AMSI tampering, LSASS dumping, LOLBins, injection
- Hunting C2 in encrypted traffic: JA4+/JA4X fingerprints, beaconing, DNS tunneling
- Hunting cloud-identity attacks: Entra device-code/OAuth phishing, PRT theft, CloudTrail abuse
- Purple-team validation: emulate ATT&CK with Atomic Red Team/Caldera, find detection gaps
- Triaging EVTX/Zeek/CloudTrail offline during IR without a SIEM
Technique Map
| Technique | ATT&CK | CWE | Reference | Script | |-----------|--------|-----|-----------|--------| | Hypothesis-driven hunt loop (PEAK/TaHiTI) | TA0043 | CWE-778 | references/methodology-hunt-loop.md | - | | ATT&CK v18 Detection-Strategies / Analytics mapping | TA0043 | CWE-778 | references/methodology-hunt-loop.md | scripts/coverage_matrix.py | | Detection-as-Code CI (lint + compile) | TA0043 | CWE-778 | references/methodology-hunt-loop.md | scripts/dac_validate.py | | Sysmon 15 PPL + tamper/visibility-gap | T1562.001 | CWE-693 | references/windows-endpoint-hunting.md | scripts/sysmon_config_2025.xml | | ETW / AMSI in-memory patch detection | T1562.001, T1562.002 | CWE-693 | references/windows-endpoint-hunting.md | scripts/evtx_hunt.py | | LSASS credential-access handle hunt | T1003.001 | CWE-522 | references/windows-endpoint-hunting.md | scripts/evtx_hunt.py | | LOLBin / process-tree anomaly hunt | T1218, T1105, T1059 | CWE-78 | references/windows-endpoint-hunting.md | scripts/evtx_hunt.py | | Sigma rule + correlation engineering | TA0043 | CWE-778 | references/sigma-rule-engineering.md | scripts/dac_validate.py | | EVTX triage (Hayabusa/Chainsaw/Velociraptor) | TA0043 | CWE-778 | references/sigma-rule-engineering.md | scripts/sigma_pipeline.sh | | JA4+/JA4X C2 fingerprinting | T1071.001 | CWE-300 | references/network-c2-hunting.md | scripts/beacon_hunter.py | | Beaconing / long-conn / prevalence (RITA-style) | T1071, T1571 | CWE-940 | references/network-c2-hunting.md | scripts/beacon_hunter.py | | DNS tunneling / DGA / DoH abuse | T1071.004, T1572 | CWE-940 | references/network-c2-hunting.md | scripts/beacon_hunter.py | | Entra device-code / OAuth consent phishing | T1528, T1566 | CWE-287 | references/cloud-identity-hunting.md | scripts/entra_hunt.kql | | PRT theft / token replay | T1550.001 | CWE-522 | references/cloud-identity-hunting.md | scripts/entra_hunt.kql | | AWS CloudTrail abuse / log tampering | T1078.004, T1098, T1562.008 | CWE-269 | references/cloud-identity-hunting.md | scripts/cloudtrail_hunt.py | | Atomic Red Team / Caldera validation | TA0043 | CWE-778 | references/purple-team-validation.md | scripts/coverage_matrix.py | | Coverage matrix + ATT&CK Navigator + gap report | TA0043 | CWE-778 | references/purple-team-validation.md | scripts/coverage_matrix.py |
Quick Start
# 0. Deploy hunting telemetry baseline (Sysmon 15+, PPL self-protected)
sysmon -accepteula -i scripts/sysmon_config_2025.xml # or: sysmon -c <file> to update
# 1. Offline endpoint triage over collected EVTX (no SIEM)
python3 scripts/evtx_hunt.py /cases/host01/EVTX --min-severity medium --json host01.json
# 2. Network: hunt C2 beacons / DNS tunneling over Zeek logs (+ optional JA4 blocklist)
zeek -r capture.pcap LogAscii::use_json=T
python3 scripts/beacon_hunter.py --conn conn.log --dns dns.log \
--ja4-blocklist bad_ja4.txt --min-score 0.7
# 3. Cloud/identity: paste scripts/entra_hunt.kql into Sentinel/Defender;
# triage AWS offline:
python3 scripts/cloudtrail_hunt.py /cases/cloudtrail/ --json ct_findings.json
# 4. Detection-as-Code: lint + compile your Sigma repo for CI (fail-fast)
python3 scripts/dac_validate.py rules/ --backend splunk --pipeline sysmon --fail-on-error
./scripts/sigma_pipeline.sh rules/ build/ splunk microsoft365defender elasticsearch
# 5. Purple-team validate + measure coverage (ATT&CK v18 Navigator layer + gaps)
Invoke-AtomicTest T1003.001 -TestNumbers 1,2,3 # lab only; -Cleanup after
python3 scripts/coverage_matrix.py --rules rules/ --atomic-results atomic_results.json \
--watchlist watchlist.txt --navigator-out attack_layer.json --gaps-out gaps.csv
OPSEC & Detection (summary)
| Technique | Telemetry / IOC | Detection (Sigma / EDR) | OPSEC note |
|-----------|-----------------|--------------------------|------------|
| ETW/AMSI patch | RWX in ntdll/amsi; ScriptBlock w/ AmsiScanBuffer+VirtualProtect | Sigma AMSI/ETW patch rule; Sysmon EID 25; ETW-TI (kernel) | Userland patch defeats single source — correlate EID25 + ETW-TI + behavior |
| Sysmon kill | System 7036/7034, SysmonDrv unload, EPS drop to 0 | Visibility-gap metric on chatty hosts | Sysmon 15 is PPL; attacker kills agent instead — alert on stop/unload |
| LSASS dump | EID 10 handle to lsass + .dmp write | LSASS-access Sigma; access-mask + non-system source | Baseline your own EDR/AV SourceImage set first or you flood the SOC |
| LOLBin abuse | certutil/mshta/regsvr32 + http/decode/scrobj | LOLBin Sigma; parent→child tree anomalies | Hunt cold data first; live triage tips an EDR-aware operator |
| Beaconing | periodic intervals, uniform sizes, low prevalence | beacon_hunter.py CV<0.3; RITA; long-conn on non-interactive port | Need days of logs — small PCAPs inflate FPs |
| JA4X C2 | randomized certs sharing one JA4X (Sliver/Havoc) | JA4 segment-pivot; JA4X blocklist at TLS-terminating proxy | TLS 1.3 encrypts certs — capture at egress/proxy, don't block on FP alone |
| DNS tunneling | long/high-entropy subdomains, TXT volume, NXDOMAIN spikes | DNS entropy/volume scoring; DoH-to-public Sigma | Baseline normal long-FQDN apps (CDNs) before alerting |
| Device-code phish | deviceCode sign-in, broker/OfficeHome AppId, new ASN | entra_hunt.kql #1+#2; Elastic open rules | Baseline sanctioned device-code apps; not every device-code is evil |
| PRT theft | primaryRefreshToken from multi-geo same day | PRT KQL + LSASS/cloudAP endpoint join | Control-plane is the ONLY evidence — retain logs ≥90d before you need them |
| CloudTrail abuse | StopLogging/DeleteTrail, CreateAccessKey for others, Describe* burst | cloudtrail_hunt.py; Athena tamper query | Attackers disable logging early; enable org-wide all-region trail up front |
Deep Dives
- references/methodology-hunt-loop.md — PEAK/TaHiTI hunt loop, ATT&CK v18 Detection Strategies (
DETxxxx) & Analytics (ANxxxx) replacing legacy data sources, Detection-as-Code CI/CD (PTEFv4), visibility-gap detection. - references/windows-endpoint-hunting.md — Sysmon 15 PPL & tamper detection, ETW/AMSI in-memory patch detection (ETW-TI, EID 25, ScriptBlock), LSASS handle hunting, LOLBins & process-tree anomalies.
- references/sigma-rule-engineering.md — Sigma rule anatomy, correlation rules (value_count/temporal), pySigma/sigma-cli compile, Hayabusa/Chainsaw/Velociraptor 0.74 native-Sigma EVTX triage, AI-assisted authoring (SigmaGen/Uncoder).
- references/network-c2-hunting.md — JA4+/JA4S/JA4H/JA4X fingerprinting, Sliver/Havoc shared JA4X, RITA-style beaconing & long-connection stats, DNS tunneling/DGA/DoH, C2 framework signature cheat-sheet.
- references/cloud-identity-hunting.md — Entra device-code phishing (STORM-2372, Tycoon2FA, EvilTokens), OAuth consent abuse, PRT theft, Graph enumeration, AWS CloudTrail abuse & tampering.
- references/purple-team-validation.md — Atomic Red Team unit tests, MITRE Caldera chained emulation, ATT&CK v18-aware coverage matrix, Navigator layer generation, prioritized gap analysis.
Related Skills
ai-job-search
45.0kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
41.7kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
pptx
177.9kUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an em…
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
