Skip to main content
Free plan, no cardStart free
Security testing 8 min read Updated

SOC Analyst: triage a log file or a Windows .evtx, step by step

Open SOC Analyst, drop one log file of up to 64 MB on the upload area, and read the report: the file is scanned for malware, its format is detected, 17 deterministic detections run over it, and each finding is mapped to MITRE ATT&CK with the arithmetic that fired it. A raw Windows .evtx needs no conversion. Your log and the report are never stored, so export the report as JSON before you leave the page.

About this guide

Written for
Security and compliance reviewers

Before you start

  • Plan. SOC Analyst is on the security-scanning entitlement: Solo Pro and above, including Team and Enterprise. It is not included on Free, and the dashboard marks the card when your plan does not include it.
  • Where it is. Open the SOC Analyst card on the dashboard, or SOC Analyst in the side navigation.
  • One file per analysis, up to 64 MB. Split a larger log by time, or export only the window you are investigating.
  • Accepted formats. CSV, TSV, Excel (.xlsx and .xls), JSON, JSONL and NDJSON, syslog and other .log or .txt files, CEF, LEEF, Apache and nginx access logs, an EvtxECmd CSV export (.evtx.csv) and the raw binary Windows event log (.evtx).

Catalogue your critical assets first (optional, recommended)

Priority is weighted by how critical the affected asset is, so a finding on an asset the inventory does not know carries a priority band labelled provisional. A team owner or a single-user account can add assets; a team member sees the inventory and the priorities it produces but cannot change it.

  1. Open SOC Analyst. The Asset inventory panel sits above the upload area.
  2. Use the form Add an asset to the inventory for each critical host, account or service, and give it a criticality.
  3. Check the table: each asset shows its identifier, the other names it also matches in your logs, and its criticality.
  4. Run your analysis. Findings on catalogued assets now carry a real priority instead of a provisional one.

Analyse a file

  1. Open SOC Analyst.
  2. Drop the file on the upload area (it reads Drop a file here, or click to choose), or click the area and pick the file.
  3. Wait while the page shows Scanning for malware, parsing the corpus and running detections… The file is scanned for malware before anything reads it.
  4. Read the verdict at the top of the report, then work down it in the order below.

Read the report in the order an analyst works it

  1. Verdict: the one-line answer to whether this corpus needs attention.
  2. The findings, ranked by computed risk. Narrow them with the severity and perspective filters.
  3. One finding at a time. Why this fired — the arithmetic shows the metric, the threshold and the method, and a confidence computed as a Bayesian posterior from the rule’s prior, sensitivity and false-positive rate. Below it are the affected entities, the evidence rows, a recommended action and a red-team read of what the activity would buy an attacker next. Card numbers in evidence rows are redacted in place, for example [PAN-REDACTED:16], so the shape of the row survives and the value never reaches the page.
  4. The attack chain and the ATT&CK coverage: which techniques fired, and which could not have been seen with this data.
  5. Engine evidence grade: what this corpus can support. When the analysis engine is available, the Shift handoff section also lists what the file cannot answer and what the next analyst should do first.
  6. Blast radius and the escalation decision.
  7. Detection drafts (experimental, shown when the analysis engine is available): a rule for each finding in Sigma, KQL for Microsoft Sentinel and Defender, and SPL for Splunk, with its false-positive risk and tuning notes. Review a draft before it reaches your SIEM.

Keep the report before you leave

The report is held in memory for the life of the analysis and then evicted. It is never written to a database or to disk, so it cannot be retrieved once you leave the page or once the job expires.

  1. Click Export JSON.
  2. Your browser saves the report as soc-analysis- followed by the analysis id and .json. Keep it wherever your team keeps investigation records.

Work the incident register

An incident is the one record SOC Analyst stores. The Incidents panel, above the upload area, lists open incidents worst band first, each with its acknowledge target, and keeps closed incidents in a separate list. An empty register is the normal state.

Each open incident shows a button for every state the server allows it to move to next: Triaged, Escalated, Contained, Resolved, or Closed as a false positive. Hover a button to see what the move does to the response clocks. An incident with no further moves shows terminal.

Every priority band from P0 to P4 has an acknowledge target. Only P0, P1 and P2 have a containment target; P3 and P4 show no published target rather than a number nobody agreed to.

There is no button yet that opens an incident from a finding. Incidents are opened through the API, and appear in the panel as soon as they exist.

If the upload does not work

Every upload error starts with Could not analyse. followed by one of the messages below, quoted here exactly as the page shows them.

SOC Analyst upload messages, what each one means and what to do next
The page saysWhat it meansWhat to do
That file is empty.The file has no content.Check the export that produced it, then upload it again.
That file is (its size). The limit is 64 MB — split it or upload the relevant window.The file is over the 64 MB limit.Split the log by time, or export only the window you are investigating.
The analysis job is no longer available. The corpus and the report are not stored, so a report cannot be retrieved once it has expired.The job expired. The report was never stored, so it cannot be recovered.Upload the file again. Any incident opened from it is still in the register.
The analysis is taking longer than expected and polling has stopped. The job may still finish — reload and check the jobs list.The page stopped checking; the job itself may still finish.Reload the page and check the jobs list.
Lost contact with the analysis service: (detail)The page lost its connection to the service while the analysis was running.Upload the file again in a minute. If it keeps happening, check the status page.
Could not reach the analysis service: (detail)The upload never reached the service.Check your connection and try again. If it keeps happening, check the status page.

Windows event logs (.evtx)

  • Upload the raw .evtx copied off the machine; it needs no conversion. The EvtxECmd CSV export (.evtx.csv) is still accepted.
  • A record the decoder cannot fully read is kept, marked partial and counted on every run, never guessed. It keeps its channel, event ID and timestamp, so it still appears on the timeline.
  • On our own validation run of 1,961 real Windows event records, 86.9% decoded in full. The decoder article gives the method and the two defects the run found.

Questions people ask about this

How long is a SOC Analyst report kept?
It is not kept. The corpus and the report exist in memory for the life of the analysis and are then evicted, with no database row and no file on disk, so export the report as JSON before you leave the page. The one stored record is an incident, once opened.
Who can add assets to the inventory?
A team owner or a single-user account. A team member sees the inventory and the priorities it produces but cannot change it, because an asset’s criticality decides every priority band it touches.
Why does a finding say its priority is provisional?
Because the asset involved is not in the inventory yet, so its criticality is unknown. Add it with the form in the Asset inventory panel and the next analysis gives it a real priority.
Does SOC Analyst run Sigma rules?
Not for detection. It runs 17 deterministic detections of its own, each showing its arithmetic. A Sigma or YARA pack you upload is recognised as rules rather than events, and, when the analysis engine is available, the report drafts an experimental rule for each finding in Sigma, KQL and SPL.
How do I open an incident from a finding?
There is no button for that yet. Incidents are opened through the API and appear in the Incidents panel as soon as they exist, where each one can be moved to the next state the server allows.

Sources

  1. MITRE ATT&CK, the knowledge base of adversary tactics and techniques that every SOC Analyst finding is mapped to read
  2. SigmaHQ, the generic signature format for SIEM systems, one of the three dialects the experimental detection drafts are written in read
  3. Microsoft Learn, Windows Event Log, the API behind the .evtx files the decoder reads read

Source: /documentation/soc-analyst-log-triage, updated 2026-11-15. Answers are checked against the shipped product on that date.

Found something out of date? The people who build the module answer the contact form on the home page, and corrections to this guide are dated above. Tell us.

This guide is about SOC Analyst. Twenty tools, including ours, are compared by job.

Try it against your own app.

Record one flow and get four kinds of evidence back. Judge the results yourself.

  • Free plan, no card
  • 14-day Solo Pro trial
  • Demo replies within one business day
  • Data hosted in India