building-attack-pattern-library-from-cti-reports

Files (4)

• LICENSE
• SKILL.md
• api-reference.md · references/
• agent.py · scripts/

SKILL.md (first lines)

# Building Attack Pattern Library from CTI Reports

## Overview

Cyber threat intelligence (CTI) reports from vendors like Mandiant, CrowdStrike, Talos, and Microsoft contain detailed descriptions of adversary behaviors that can be extracted, normalized, and cataloged into a structured attack pattern library. This skill covers parsing CTI reports to extract adversary techniques, mapping behaviors to MITRE ATT&CK technique IDs, creating STIX 2.1 Attack Pattern objects, building a searchable library indexed by tactic, technique, and threat actor, and generating detection rule templates from documented patterns.


## When to Use

- When deploying or configuring building attack pattern library from cti reports capabilities in your environment
- When establishing security controls aligned to compliance requirements
- When building or improving security architecture for this domain
- When conducting security assessments that require this implementation

## Prerequisites

- Python 3.9+ with `stix2`, `mitreattack-python`, `spacy`, `requests` libraries
- Collection of CTI reports (PDF, HTML, or text format)
- MITRE ATT&CK STIX data (local or via TAXII)
- Understanding of ATT&CK technique structure and naming conventions
- Familiarity with detection engineering concepts (Sigma, YARA)

## Key Concepts

### Attack Pattern Extraction

CTI reports describe adversary behaviors in natural language. Extraction involves identifying action verbs and technical terms that map to ATT&CK techniques, recognizing tool names and malware families, identifying infrastructure indicators, and mapping sequences of behaviors to attack chains (kill chain phases).

### STIX 2.1 Attack Pattern Objects

STIX defines Attack Pattern as a Structured Domain Object (SDO) that describes ways threat actors attempt to compromise targets. Each pattern links to ATT&CK via external references, includes kill chain phases (tactics), and can be related to Intrusion Sets, Malware, and Tool objects.

### Detection Rule Generation

Extracted attack patterns inform detection engineering by providing: specific procedure examples for Sigma rule creation, behavioral sequences for correlation rules, IOC patterns for YARA and Snort rules, and data source requirements for telemetry gaps.

## Workflow

### Step 1: Parse CTI Reports and Extract Behaviors

```python
import re
import json
from collections import defaultdict

class CTIReportParser:
    """Parse CTI reports to extract adversary behaviors."""

    BEHAVIOR_INDICATORS = [
        "used", "executed", "deployed", "leveraged", "exploited",
        "established", "created", "modified", "downloaded", "uploaded",
        "exfiltrated", "injected", "enumerated", "spawned", "dropped",
        "persisted", "escalated", "moved laterally", "collected",
        "encrypted", "compressed", "encoded", "obfuscated",
    ]

    TOOL_PATTERNS = [
        r'\b(Cobalt Strike|Mimikatz|PsExec|BloodHound|Rubeus|Impacket)\b',
        r'\b(PowerShell|cmd\.exe|WMI|WMIC|certutil|bitsadmin)\b',
        r'\b(Metasploit|Empire|Covenant|Sliver|Brute Ratel)\b',
        r'\b(Lazagne|SharpHound|ADFind|Sharphound|Invoke-Obfuscation)\b',
    ]

    TECHNIQUE_KEYWORDS = {
        "spearphishing": "T1566",
        "phishing attachment": "T1566.001",
        "phishing link": "T1566.002",
        "powershell": "T1059.001",
        "command line": "T1059.003",
        "scheduled task": "T1053.005",
        "registry run key": "T1547.001",
        "process injection": "T1055",
        "dll side-loading": "T1574.002",
        "credential dumping": "T1003",
        "lsass": "T1003.001",
        "kerberoasting": "T1558.003",
        "pass the hash": "T1550.002",
        "remote desktop": "T1021.001",
        "smb": "T1021.002",
        "winrm": "T1021.006",
        "data staging": "T1074",
        "exfiltration over c2": "T1041",
        "dns tunneling": "T1071.004",
        "web shell": "T1505.003",
    }

    def parse_report(self, text, report_metadata=None):
        """Parse a CTI report and extract behaviors."""
        sentences = re.split(r'[.!?]\s+', text)
        behaviors = []

        for sentence in sentences:
            sentence_lower = sentence.lower()
            # Check for behavior indicators
            for indicator in self.BEHAVIOR_INDICATORS:
                if indicator in sentence_lower:
                    behavior = {
                        "sentence": sentence.strip(),
                        "action": indicator,
                        "tools": self._extract_tools(sentence),
                        "technique_hints": self._match_techniques(sentence_lower),
                    }
                    if behavior["technique_hints"]:
                        behaviors.append(behavior)
                    break

        print(f"[+] Extracted {len(behaviors)} behavioral indicators from report")
        return behaviors

    def _extract_tools(self, text):
        """Extract tool/malware names from text."""
        tools = set()
        for pattern in self.TOOL_PATTERNS:
            matches = re.findall(pattern, text, re.IGNORECASE)
            tools.update(matches)
        return list(tools)

    def _match_techniques(self, text):
        """Match text to ATT&CK technique hints."""
        matches = []

From mukul975/Anthropic-Cybersecurity-Skills, Apache-2.0.

More from mukul975/Anthropic-Cybersecurity-Skills