Extracting and writing up IOCs
Turning analysis into something others can act on: choosing which indicators are worth sharing, defanging, adding context and expiry, hunting them across the estate, expressing them in STIX 2.1 and MISP, and writing a malware triage report that separates fact from inference.
01Which indicators to extract
- Select useful indicators and drop the noise
- Write each indicator with context, confidence and expiry
- Hunt indicators across endpoints, network and email
- Share in STIX 2.1 / MISP and write a clear triage report
A sample can yield hundreds of observables. Most are noise: Windows DLLs it loads, Microsoft domains it contacts for certificate checks, generic user agents. Extract the ones that are specific to this threat and useful to someone: a SOC that can block, an EDR that can hunt, a partner who can check their logs.
| Indicator | Keep when | Drop when |
|---|---|---|
| File hashes (SHA-256 of sample, drops, second stages) | Always, for scoping | Hashes of legitimate system files the sample used |
| Domains and URLs | Attacker-registered or compromised C2 and download paths | CDN, cloud provider or connectivity-check domains |
| IP addresses | Dedicated C2 servers | Shared hosting, cloud front-ends, sinkholes |
| Host artefacts (mutex, file path, Run value, service name) | Distinctive to the family | Generic paths like %TEMP% |
| Behaviours (remote thread into argument-less rundll32, ping/del self-delete) | Always; they become detections | - |
Before any network indicator goes to a blocklist, check it against known-good infrastructure and your own traffic. Blocking a cloud provider IP that the sample happened to use can take down business services.
This is an advanced module
Malware analysis for responders is part of the Pro pass. 4 more sections — the deep dives, the final design, follow-ups and the test — unlock with it.
Already a cyart.io or studio.cyart.io member? Your pass code is in your Studio account.