Complete Claude skills package covering every phase of malware analysis from initial triage to production-ready detection rules.
π‘ Want to learn the manual analysis techniques behind these skills? Check out the comprehensive Malware Analysis SOP covering traditional analysis methods and tools.
5 Specialized Skills covering the complete malware analysis lifecycle:
- malware-triage - Initial assessment and prioritization
- malware-dynamic-analysis - Safe execution and behavior monitoring
- specialized-file-analyzer - Non-PE file analysis (.NET, Office, PDF, scripts)
- detection-engineer - Detection rules and IOC management
- malware-report-writer - Professional report creation
Not sure which skill to use? Use this decision tree:
START: What do you need to do?
ββ "I need to quickly assess a sample" β Use: malware-triage
β
ββ "I need to execute malware safely and monitor it" β Use: malware-dynamic-analysis
β
ββ "I have a document/script/non-executable" β Use: specialized-file-analyzer
β ββ .NET/C# assembly (.exe with .NET)
β ββ Office document with macros (.docm, .xlsm)
β ββ PDF file
β ββ PowerShell/VBS/JavaScript
β ββ Archive (.zip, .rar, .7z)
β ββ Linux binary (ELF)
β ββ Shortcut file (.lnk)
β
ββ "I need to create detection rules or defang IOCs" β Use: detection-engineer
β ββ YARA rules β Use: malware-report-writer
β ββ Sigma rules (SIEM) β Use: detection-engineer
β ββ Suricata rules (IDS) β Use: detection-engineer
β ββ Defang IOCs β Use: detection-engineer
β
ββ "I need to write a malware analysis report" β Use: malware-report-writer
| Skill Name | When to Use | Primary Output |
|---|---|---|
| malware-triage | First look at unknown sample | Classification, priority, initial IOCs |
| malware-dynamic-analysis | Execute and monitor behavior | Behavioral IOCs, process tree, network traffic |
| specialized-file-analyzer | Non-PE files (docs, scripts, etc.) | Deobfuscated code, embedded payloads, IOCs |
| detection-engineer | Create detection rules, defang IOCs | Sigma/Suricata rules, hunting queries |
| malware-report-writer | Document findings professionally | Complete technical report, YARA rules |
-
Upload Skills to Claude
- Open Claude (claude.ai)
- Click paperclip icon (π)
- Upload all 5
.skillfolders or individualSKILL.mdfiles - Skills install automatically
-
Verify Installation
Ask Claude: "What skills do you have?" Should list: malware-triage, malware-dynamic-analysis, etc. -
Start Analyzing
"Help me triage this suspicious .exe file" "Guide me through safely executing this malware" "Analyze this Office document with macros"
Problem: Malware analysis VMs should be isolated with NO internet, but Claude Code requires internet.
Solution: Run Claude Code on your HOST machine, analyze exported evidence from isolated VMs.
βββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β HOST MACHINE (Internet-connected) β
β β’ Claude Code running here β
β β’ Skills loaded and ready β
β β’ Analyzes exported evidence β
β β’ Creates reports and detection rules β
ββββββββββββββββββββ¬βββββββββββββββββββββββββββββββββββ
β
β Evidence Transfer:
β β’ Shared folder (read-only)
β β’ USB drive
β β’ Isolated network file share
β
ββββββββββββββββββββΌβββββββββββββββββββββββββββββββββββ
β ANALYSIS VM (Isolated - NO Internet) β
β REMnux / FlareVM β
β β’ Execute malware here β
β β’ Run monitoring tools β
β β’ Capture evidence β
β β’ Export results only β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββ
Set up on your HOST machine:
/malware-analysis/
βββ samples/ # Store original samples (DO NOT share with VM)
βββ evidence/ # Evidence exported FROM analysis VM
β βββ strings/ # strings output
β βββ procmon/ # Process Monitor logs (CSV)
β βββ wireshark/ # Network captures (text exports)
β βββ sysmon/ # Sysmon event logs (JSON/CSV)
β βββ memory/ # Memory dump analysis results
β βββ screenshots/ # Behavior screenshots
βββ analysis/ # Claude Code analysis workspace
β βββ notes.md
β βββ findings.md
βββ detections/ # Detection rules created by Claude Code
β βββ yara/
β βββ sigma/
β βββ suricata/
βββ reports/ # Final deliverables
βββ malware_report.mdOn Analysis VM (REMnux/FlareVM):
# Configure shared folder or USB mount point
mkdir /mnt/evidence_export
# Make it write-only if possibleOn Host Machine:
# Create analysis structure
mkdir -p /malware-analysis/{evidence,analysis,detections,reports}
cd /malware-analysis# Use malware-triage skill
"Help me prepare for analyzing a suspected ransomware sample"
"What Procmon filters should I configure for ransomware?"
"What Wireshark display filters should I use for C2 traffic?"
"What Sysmon events should I focus on for persistence?"Claude Code will guide you through:
- Tool configuration commands
- Filter setup
- Monitoring checklist
In your Analysis VM terminal:
# 1. Take snapshot
virsh snapshot-create malware-analysis-vm
# 2. Start monitoring tools
# - Procmon (with filters from Claude)
# - Wireshark
# - Process Hacker
# - Regshot (1st shot)
# 3. Execute malware
./malware.exe
# 4. Monitor for 15+ minutes
# 5. Stop all captures
# 6. Export evidence to shared folder
cp procmon_filtered.csv /mnt/evidence_export/
cp wireshark_http.txt /mnt/evidence_export/
wevtutil epl Microsoft-Windows-Sysmon/Operational /mnt/evidence_export/sysmon.evtx
# Convert to CSV or JSON for easier analysis
# 7. Take screenshots of any unusual behavior
# 8. Revert to snapshot
virsh snapshot-revert malware-analysis-vm <snapshot-name>Now use Claude Code with your skills:
4a. Triage the Evidence:
cd /malware-analysis/evidence
# In Claude Code:
"Triage this sample based on the evidence I've collected"
Read procmon/malware_activity.csv
Read wireshark/http_traffic.txt
Read sysmon/events.json
"What type of malware is this?"
"What are the key IOCs?"
"What persistence mechanisms were used?"4b. Deep Analysis:
# Use specialized-file-analyzer if needed
"Analyze this PowerShell script extracted from memory"
Read evidence/memory/decoded_script.ps1
# Use malware-dynamic-analysis skill
"Analyze this Procmon output for ransomware encryption behavior"
"What C2 infrastructure was contacted based on this network capture?"4c. Create Detections:
# Use detection-engineer skill
"Create Sigma rules for this malware based on the Sysmon events"
Read evidence/sysmon/events.csv
"Create a Suricata rule for this C2 traffic"
Read evidence/wireshark/c2_communication.txt
"Defang all these IOCs for my report"
# Claude will create rules in detections/ folder4d. Write Report:
# Use malware-report-writer skill
"Create a complete malware analysis report based on my findings"
"Include these Procmon behaviors: [paste key findings]"
"Network indicators: [paste from Wireshark]"
"Persistence mechanisms: [paste from Sysmon]"
# Claude will generate the full report in reports/Analysis of Tool Outputs:
- Read and interpret Procmon CSV exports
- Analyze Wireshark text exports (HTTP objects, DNS queries)
- Parse Sysmon event logs (CSV/JSON)
- Interpret strings output
- Analyze PE header info from tools
- Review memory dump analysis results
Tool Setup Guidance:
- Procmon filter configuration
- Wireshark display filters
- Sysmon config recommendations
- INetSim/FakeNet setup
- VM isolation verification
Detection Creation:
- YARA rules from your findings
- Sigma rules from Sysmon events
- Suricata rules from network captures
- Hunting queries for EDR/SIEM
- IOC defanging and formatting
Report Writing:
- Complete malware analysis reports
- Executive summaries
- Technical documentation
- IOC lists with context
- Remediation recommendations
Execute Malware:
- Won't run
./malware.exe - Won't help improve malware code
- Won't make malware more evasive
Read Raw Binaries Directly:
- Won't directly analyze PE/ELF files
- Use tools first (strings, objdump, etc.)
- Then have Claude analyze the output
# ===== On Analysis VM (Manual) =====
analyst@remnux:~$ strings malware.exe > /mnt/export/strings.txt
analyst@remnux:~$ objdump -d malware.exe > /mnt/export/disasm.txt
analyst@remnux:~$ # Execute malware with monitoring
analyst@remnux:~$ cp evidence/* /mnt/export/
# ===== On Host (Claude Code) =====
you@host:~$ cd /malware-analysis/evidence
you@host:~$ # In Claude Code session:
> "Help me triage this sample"
> Read strings.txt
> Read procmon_export.csv
Claude: "Based on the evidence, this appears to be a remote access
trojan (RAT). Key indicators:
- Network connection to 185.220.101.x:4444
- Registry persistence via Run key
- Process injection into explorer.exe
- Keylogging behavior (GetAsyncKeyState calls)
..."
> "Create detection rules for this RAT"
Claude: [Creates YARA rule, Sigma rule, Suricata rule]
> "Write the complete analysis report"
Claude: [Generates full professional report]-
Export Evidence in Text Formats:
- Procmon β CSV (not .PML)
- Wireshark β Text export, HTTP objects
- Sysmon β CSV or JSON
- Memory analysis β Text output
-
Use Read-Only Shares:
- VM can only write to share, not read
- Prevents VM from accessing your host files
-
Batch Analysis:
- Analyze multiple samples
- Export all evidence
- Use Claude Code to batch analyze
-
Templates and Automation:
- Create export scripts in VM
- Use Claude Code to create analysis templates
- Streamline your workflow
-
Version Control:
- Keep analysis notes in git
- Track detection rule versions
- Version your reports
Q: Claude Code refuses to read my evidence file A: Ensure it's in text format (CSV, JSON, TXT), not binary (PML, PCAP, EVTX)
Q: How do I convert PCAP to text? A: Use tshark:
tshark -r capture.pcap -Y http -T fields -e http.host -e http.request.uri > http_traffic.txtQ: How do I convert EVTX to CSV? A: On Windows:
Get-WinEvent -Path sysmon.evtx | Export-Csv sysmon.csvQ: Claude Code says file is too large A: Filter or summarize first:
# Instead of reading all 50k lines
grep "malware.exe" procmon_huge.csv > procmon_filtered.csv
# Then read the filtered versionPurpose: Rapid initial assessment (5-30 minutes per sample)
Use When:
- You have a new unknown sample
- Need to prioritize multiple samples
- Want quick classification before deep analysis
- Time-constrained analysis
Provides:
- File hashing (MD5, SHA1, SHA256)
- Online reputation lookup guidance
- PE structure analysis
- String extraction workflow
- Malware type classification
- Threat level assessment
- Behavior prediction
- Priority determination
- Triage report template
Key Features:
- Structured 5-phase workflow
- Suspicious API import database
- Classification decision tree
- Time management strategies
- Integration with full analysis
Files:
malware-triage/
βββ SKILL.md
βββ references/
β βββ indicators.md # Suspicious APIs & patterns
β βββ triage_checklist.md # Step-by-step guide
βββ scripts/
βββ hash_calculator.py
Purpose: Safe execution and comprehensive behavior monitoring
Use When:
- Need to observe actual malware behavior
- Validating static analysis hypotheses
- Capturing C2 communications
- Documenting runtime activity
- Extracting dynamic IOCs
Provides:
- Pre-execution safety checklist
- Tool setup (Procmon, Wireshark, Process Hacker, Sysmon)
- Monitoring workflows for:
- Process activity and injection
- File system operations
- Registry modifications
- Network communications
- Persistence mechanisms
- Artifact collection procedures
- Automated sandbox usage (ANY.RUN, Joe Sandbox)
- Evidence export workflows
Key Features:
- Safety-first approach with isolation verification
- Step-by-step tool configuration
- Real-time monitoring guidance
- Behavioral IOC extraction
- Timeline creation
- Integration with report writing
Files:
malware-dynamic-analysis/
βββ SKILL.md
βββ references/
β βββ tool_setup.md # Procmon, Wireshark, etc.
β βββ sandbox_setup.md # Local sandbox installation
β βββ anti_analysis_bypass.md # VM detection bypass
βββ scripts/
βββ (monitoring automation scripts)
Purpose: Analyze non-PE file formats requiring specialized tools
Use When You Encounter:
- .NET/C# Assemblies - Managed executables requiring decompilation
- Office Documents - Word/Excel with VBA/XLM macros
- PDF Files - Suspicious PDFs with JavaScript/exploits
- Scripts - PowerShell, VBScript, JavaScript, Batch files
- Archives - ZIP, RAR, 7z, TAR.GZ
- Shortcuts - .LNK files
- Linux Binaries - ELF executables
Provides:
- Per-Format Workflows:
- .NET: dnSpy/ILSpy decompilation, de4dot deobfuscation
- Office: oledump/olevba macro extraction, XLM macro analysis
- PDF: pdfid/pdf-parser/peepdf JavaScript extraction
- PowerShell: PSDecode deobfuscation, pattern analysis
- Scripts: Deobfuscation techniques
- Archives: Safe inspection and extraction
- LNK: LECmd/lnkinfo analysis
- ELF: readelf/strace/ltrace analysis
Key Features:
- Quick file type identification
- Format-specific tooling
- Deobfuscation techniques
- IOC extraction per format
- Integration with main report
Files:
specialized-file-analyzer/
βββ SKILL.md
βββ references/
β βββ dotnet_analysis.md
β βββ office_macros.md
β βββ pdf_analysis.md
β βββ script_analysis.md
βββ scripts/
βββ (format-specific helpers)
Purpose: Convert analysis findings into production detection rules
Use When You Need:
- Sigma Rules for SIEM (Splunk, Elastic, QRadar, Sentinel)
- Suricata/Snort Rules for network IDS/IPS
- Hunting Queries for SOC/threat hunting
- IOC Defanging for safe documentation
- IOC Format Conversion (CSV, STIX, OpenIOC)
Provides:
- IOC defanging workflows (URLs, IPs, domains, emails)
- IOC confidence & volatility assessment
- Sigma rule creation:
- Process creation detection
- Network connection monitoring
- Registry modification detection
- File creation alerts
- Logsource configuration
- Sigma-to-SIEM conversion
- Suricata rule creation:
- HTTP/HTTPS C2 detection
- DNS query monitoring
- TLS/certificate analysis
- Malware download detection
- Hunting query templates (Splunk, Elastic, EDR)
- IOC export formats (STIX 2.1, CSV, OpenIOC)
Key Features:
- Automated IOC defanging
- Rule testing procedures
- False positive analysis
- MITRE ATT&CK tagging
- Multi-format export
Files:
detection-engineer/
βββ SKILL.md
βββ references/
β βββ sigma_examples.md
β βββ suricata_examples.md
β βββ ioc_standards.md
βββ scripts/
βββ ioc_defanger.py
Purpose: Create professional, enterprise-ready malware analysis reports
Use When:
- Documenting analysis findings
- Creating formal deliverables
- Writing for executive/technical audiences
- Formalizing IOCs and detection rules
- Creating YARA signatures
Provides:
- Complete report template (12 sections)
- Executive summary guidelines
- Technical analysis structure
- IOC documentation standards
- YARA rule creation (testing + validation)
- Remediation recommendations
- Quality assurance checklist
- Efficient workflow strategies
Key Features:
- Multi-audience writing guidance
- YARA rule best practices
- Common mistakes to avoid
- Section-by-section templates
- Professional formatting
Files:
malware-report-writer/
βββ SKILL.md
βββ assets/
β βββ report_template.md # Complete report structure
βββ references/
βββ best_practices.md # Writing standards & tips
Skill: malware-triage
- Calculate hashes for all samples
- Quick reputation lookups (VirusTotal, MalwareBazaar)
- Rapid classification
- Prioritize samples (High/Medium/Low)
- Document initial findings
Output: Triage reports for all samples, priority list
For Standard PE Executables:
Skill: malware-dynamic-analysis
- Setup isolated VM environment
- Configure monitoring tools (Procmon, Wireshark, Process Hacker)
- Execute malware
- Monitor behavior (15-60 minutes)
- Extract behavioral IOCs
- Collect artifacts
For Non-PE Files:
Skill: specialized-file-analyzer
- Identify file type (
filecommand) - Select appropriate analysis tool:
- .NET β dnSpy/ILSpy
- Office β oledump/olevba
- PDF β pdfid/pdf-parser
- Scripts β PSDecode/manual analysis
- ELF β readelf/strace
- Extract/deobfuscate content
- Document functionality
- Extract IOCs
Output: Complete technical analysis, behavioral IOCs, screenshots, PCAPs
Skill: detection-engineer
- Review all extracted IOCs
- Defang IOCs for documentation
- Assess IOC confidence levels
- Create Sigma rules for behavioral detection
- Create Suricata rules for network detection
- Generate hunting queries
- Test all rules
Output: Tested Sigma rules, Suricata rules, defanged IOC list, hunting queries
Skills: malware-report-writer + detection-engineer
-
Draft technical sections
- Sample information
- Static analysis findings
- Dynamic analysis observations
- Specialized file analysis results
-
Add detection content
- YARA rules (test against sample!)
- Sigma rules
- Suricata rules
- Defanged IOCs with confidence ratings
- Remediation recommendations
-
Write non-technical sections
- Executive summary
- Malware classification
- Conclusion
- Appendices
-
Quality assurance
- Run through quality checklist
- Test all detection rules
- Verify all hashes
- Grammar/spelling check
- Final formatting
Output: Professional malware analysis report (PDF)
User: "I need to analyze this suspicious .exe sample"
Workflow:
1. "Help me triage this sample" β malware-triage
- Calculates hashes
- Checks VirusTotal
- Analyzes PE structure
- Classifies as trojan/RAT
- Priority: High
2. "Guide me through dynamic analysis" β malware-dynamic-analysis
- Verifies VM isolation
- Sets up Procmon, Wireshark, Process Hacker
- Executes sample
- Monitors for 15 minutes
- Extracts behavioral IOCs
3. "Create detection rules for this malware" β detection-engineer
- Defangs all IOCs
- Creates Sigma rule for process injection
- Creates Suricata rule for C2 traffic
- Generates hunting queries
4. "Help me write the analysis report" β malware-report-writer
- Uses report template
- Structures all findings
- Creates YARA rule
- Reviews with quality checklist
User: "I have a suspicious Word document with macros"
Workflow:
1. "Help me triage this .docm file" β malware-triage
- Identifies Office document
- Checks VirusTotal
- Classifies as macro dropper
2. "Analyze the macros in this document" β specialized-file-analyzer
- Guides oledump.py usage
- Extracts VBA macro code
- Identifies AutoOpen function
- Deobfuscates strings
- Finds PowerShell download cradle
3. "Create detections for this macro malware" β detection-engineer
- Defangs URLs from macro
- Creates Sigma rule for macro execution
- Creates hunting query for similar documents
4. "Write the report" β malware-report-writer
- Documents macro analysis
- Includes sanitized macro code
- Creates YARA rule for document
User: "I have 10 samples to analyze, help me prioritize"
Workflow:
1. "Triage all 10 samples" β malware-triage (use 10 times)
- Quick triage (5-10 min each)
- Hash + VirusTotal for all
- Classifications recorded
- Priority assigned
2. Results:
- 3 High Priority (unknown/sophisticated)
- 5 Medium Priority (known variants)
- 2 Low Priority (common adware)
3. Strategy:
- Phase 1-2: Deep analysis on 3 high-priority samples
- Phase 2-3: Quick analysis on 5 medium-priority samples
- Document low-priority findings (note in report)
- β Triage ALL samples first (prioritize effectively)
- β Document as you analyze (don't leave for the end)
- β Create detection rules during analysis (not as an afterthought)
- β Test YARA rules immediately (avoid failures later)
- β Don't spend excessive time on one sample (spread effort wisely)
- β Better to deeply analyze 2-3 samples than shallowly analyze 10
- β Focus on creating working detection rules
- β Document methodology clearly
- β Don't submit untested YARA/Sigma rules
- β Triage skill for ALL samples (fast initial assessment)
- β Dynamic analysis for unknowns (need behavioral data)
- β Specialized file analyzer when needed (don't force PE analysis on docs)
- β Detection engineer at the end (consolidate IOCs)
- β Report writer continuously (update as you go)
- Use quick triage (5 min) for initial all-sample sweep
- Use standard triage (15 min) for high-priority samples
- Don't skip hash calculation (required for all samples)
- Document predicted behaviors (guides dynamic analysis)
- ALWAYS verify VM isolation before execution
- Start monitoring tools BEFORE executing malware
- Observe for minimum 15 minutes (some malware delays)
- Export ALL artifacts before reverting VM
- Take screenshots of unusual behavior
- File type first - use
filecommand to confirm - Right tool for format - don't use PE tools on Office docs
- Test deobfuscation output (ensure it makes sense)
- Save both obfuscated and deobfuscated versions
- Document deobfuscation methodology
- Always defang IOCs in reports (avoid accidental clicks)
- Test rules before documenting (must work!)
- Label IOC confidence (High/Medium/Low)
- Note IOC volatility (Static/Dynamic)
- Include false positive analysis
- Executive summary last (after understanding full scope)
- Test all YARA rules against sample AND clean files
- Use quality checklist before submission
- Grammar/spelling matters (professional appearance)
- Include all three hashes (MD5, SHA1, SHA256)
- TCM Security: Practical Malware Analysis & Triage (https://certifications.tcm-sec.com/pmrp/)
- SANS: FOR610, FOR710 malware analysis courses
- Open Security Training: Free malware analysis courses
- MalwareBazaar: https://bazaar.abuse.ch/
- VirusTotal: https://www.virustotal.com/
- Hybrid Analysis: https://www.hybrid-analysis.com/
- Sysinternals Suite: https://docs.microsoft.com/en-us/sysinternals/
- Wireshark: https://www.wireshark.org/docs/
- dnSpy: https://github.com/dnSpy/dnSpy
- Didier Stevens Tools: https://blog.didierstevens.com/programs/
- Sigma: https://github.com/SigmaHQ/sigma
- Suricata: https://docs.suricata.io/
- SANS Malware Analysis Cheat Sheet: https://www.sans.org/posters/
- MITRE ATT&CK: https://attack.mitre.org/
- MalAPI.io (Windows APIs): https://malapi.io/
Q: Do I need to use all 5 skills? A: No! Use skills based on your specific needs. Triage and report-writer are most common. Others are situational.
Q: Can I use these skills for real-world work? A: Absolutely! These skills are based on enterprise malware analysis standards and workflows.
Q: Which skill should I start with?
A: Always start with malware-triage for initial assessment, then branch to others as needed.
Q: How do I know if a file needs specialized-file-analyzer?
A: Run file sample.bin. If output shows: .NET assembly, Office document, PDF, script, or ELF β use specialized-file-analyzer.
Q: Are these skills suitable for professional work? A: Yes! These skills are based on industry-standard malware analysis workflows used by security professionals and incident responders.
Q: Can I modify skill prompts? A: Yes! Customize to fit your workflow and preferences.
Q: What if I encounter a file type not covered? A: Use the general principles from malware-triage and dynamic-analysis, then search for format-specific tools.
Before finalizing your analysis, verify:
Analysis Phase:
- All samples triaged (hashes, classification, priority)
- High-priority samples deeply analyzed (static + dynamic)
- Specialized files analyzed with appropriate tools
- All artifacts collected (PCAPs, screenshots, logs)
- Behavioral IOCs extracted
Detection Phase:
- All IOCs defanged properly
- IOC confidence levels assigned
- YARA rules created AND TESTED
- Sigma rules created and validated
- Suricata rules tested (if applicable)
- Hunting queries documented
Report Phase:
- Report uses professional template
- Executive summary written (non-technical)
- All technical sections complete
- All three hashes included for each sample
- Detection rules tested and working
- Grammar/spelling checked
- Quality checklist completed
- Ready for delivery!
| Your Question | Use This Skill |
|---|---|
| "What is this file?" | malware-triage |
| "How do I execute this safely?" | malware-dynamic-analysis |
| "How do I analyze this Word doc?" | specialized-file-analyzer |
| "How do I analyze this .NET exe?" | specialized-file-analyzer |
| "How do I analyze this PowerShell script?" | specialized-file-analyzer |
| "How do I create a Sigma rule?" | detection-engineer |
| "How do I defang IOCs?" | detection-engineer |
| "How do I write the report?" | malware-report-writer |
| "How do I create a YARA rule?" | malware-report-writer |
This complete toolkit provides everything needed for professional malware analysis:
- β 5 specialized skills covering full analysis lifecycle
- β Expert-level workflows for each analysis phase
- β Clear skill selection guidance (use the right tool for the job)
- β Integration strategies between skills
- β Industry best practices and time management
- β Quality assurance checklists for professional deliverables
Upload the skills and start analyzing!
Built for security professionals, incident responders, and malware analysts. ππ―