# Analyzing Threat Intelligence Feeds

> Ingests, normalizes, and enriches structured and unstructured threat intelligence feeds into STIX 2.1 format, evaluating feed quality and deduplicating indicators for distribution to SIEM, firewall, and EDR platforms.

- Skill: `mukul975/analyzing-threat-intelligence-feeds` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds add mukul975/analyzing-threat-intelligence-feeds`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mukul975/analyzing-threat-intelligence-feeds/raw
- Safety review: CAUTION (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security, Data & Analytics, Marketing & Growth, ETL & Pipelines, Incident Response
- Tags: Ioc, Misp, Opencti, Recorded Future, Stix, Taxii, Threat Intelligence, Threatconnect
- License: Apache-2.0
- Author: mukul975 (https://skillmd.com/u/mukul975)
- Updated: 2026-07-06
- Page: https://skillmd.com/skills/mukul975/analyzing-threat-intelligence-feeds

---

# Analyzing Threat Intelligence Feeds

## When to Use

Use this skill when:
- Ingesting new commercial or OSINT threat feeds and assessing their signal-to-noise ratio
- Normalizing heterogeneous IOC formats (STIX 2.1, OpenIOC, YARA, Sigma) into a unified schema
- Evaluating feed freshness, fidelity, and relevance to the organization's threat profile
- Building automated enrichment pipelines that correlate IOCs against SIEM events

**Do not use** this skill for raw packet capture analysis or live incident triage without first establishing a CTI baseline.

## Prerequisites

- Access to a Threat Intelligence Platform (TIP) such as ThreatConnect, MISP, or OpenCTI
- API keys for at least one commercial feed (Recorded Future, Mandiant Advantage, or VirusTotal Enterprise)
- TAXII 2.1 client library (taxii2-client Python package or equivalent)
- Role with read/write permissions to the TIP's indicator database

## Workflow

### Step 1: Enumerate and Prioritize Feed Sources

List all available feeds categorized by type (commercial, government, ISAC, OSINT):
- Commercial: Recorded Future, Mandiant Advantage, CrowdStrike Falcon Intelligence
- Government: CISA AIS (Automated Indicator Sharing), FBI InfraGard, MS-ISAC
- OSINT: AlienVault OTX, Abuse.ch, PhishTank, Emerging Threats

Score each feed on: update frequency, historical accuracy rate, coverage of your sector, and attribution depth. Use a weighted scoring matrix with criteria from NIST SP 800-150 (Guide to Cyber Threat Information Sharing).

### Step 2: Ingest via TAXII 2.1 or API

For TAXII-enabled feeds:
```
taxii2-client discover https://feed.example.com/taxii/
taxii2-client get-collection --collection-id <id> --since 2024-01-01
```

For REST API feeds (e.g., Recorded Future):
- Query `/v2/indicator/search` with `risk_score_min=65` to filter low-confidence IOCs
- Apply rate limiting and exponential backoff for API resilience

### Step 3: Normalize to STIX 2.1

Convert each IOC to STIX 2.1 objects using the OASIS standard schema:
- IP address → `indicator` object with `pattern: "[ipv4-addr:value = '...']"`
- Domain → `indicator` with `pattern: "[domain-name:value = '...']"`
- File hash → `indicator` with `pattern: "[file:hashes.SHA-256 = '...']"`

Attach `relationship` objects linking indicators to `threat-actor` or `malware` objects. Use `confidence` field (0–100) based on source fidelity rating.

### Step 4: Deduplicate and Enrich

Run deduplication against existing TIP database using normalized value + type as composite key. Enrich surviving IOCs:
- VirusTotal: detection ratio, sandbox behavior reports
- PassiveTotal (RiskIQ): WHOIS history, passive DNS, SSL certificate chains
- Shodan: banner data, open ports, geographic location

### Step 5: Distribute to Consuming Systems

Export enriched indicators via TAXII 2.1 push to SIEM (Splunk, Microsoft Sentinel), firewalls (Palo Alto XSOAR playbooks), and EDR platforms. Set TTL (time-to-live) per indicator type: IP addresses 30 days, domains 90 days, file hashes 1 year.

## Key Concepts

| Term | Definition |
|------|-----------|
| **STIX 2.1** | Structured Threat Information Expression — OASIS standard JSON schema for CTI objects including indicators, threat actors, campaigns, and relationships |
| **TAXII 2.1** | Trusted Automated eXchange of Intelligence Information — HTTPS-based protocol for sharing STIX content between servers and clients |
| **IOC** | Indicator of Compromise — observable artifact (IP, domain, hash, URL) that indicates a system may have been breached |
| **TLP** | Traffic Light Protocol — color-coded classification (RED/AMBER/GREEN/WHITE) defining sharing restrictions for CTI |
| **Confidence Score** | Numeric value (0–100 in STIX) reflecting the producer's certainty about an indicator's malicious attribution |
| **Feed Fidelity** | Historical accuracy rate of a feed measured by true positive rate in production detections |

## Tools & Systems

- **ThreatConnect TC Exchange**: Aggregates 100+ commercial and OSINT feeds; provides automated playbooks for IOC enrichment
- **MISP (Malware Information Sharing Platform)**: Open-source TIP supporting STIX/TAXII; widely used by ISACs and government CERTs
- **OpenCTI**: Open-source platform with native MITRE ATT&CK integration and graph-based relationship visualization
- **Recorded Future**: Commercial feed with AI-powered risk scoring and real-time dark web monitoring
- **taxii2-client**: Python library for TAXII 2.0/2.1 client operations (pip install taxii2-client)
- **PyMISP**: Python API for MISP feed management and IOC submission

## Common Pitfalls

- **IOC age staleness**: IP addresses and domains rotate frequently; applying 1-year-old IOCs generates false positives. Enforce TTL policies.
- **Missing context**: Blocking an IOC without understanding the associated campaign or adversary can disrupt legitimate business traffic (e.g., CDN IPs shared with malicious actors).
- **Feed overlap without deduplication**: Ingesting the same IOC from five feeds without deduplication inflates indicator counts and SIEM rule complexity.
- **TLP violation**: Redistributing RED-classified intelligence outside authorized boundaries violates sharing agreements and trust relationships.
- **Over-blocking on low-confidence indicators**: Indicators with confidence below 50 should trigger detection-only rules, not blocking, to avoid operational disruption.