The ThreatOps Skills Map: Eight Abilities and a Ninety-Day Learning Route

The ThreatOps Skills Map: Eight Abilities and a Ninety-Day Learning Route

Operations professionals regularly ask which security specialism is worth building next. One answer that deserves more attention is ThreatOps: the work of turning outside intelligence about attackers into regular, evidence-backed searches of your own environment. It sits between analysis and engineering, which makes it a useful target for people who like both reading and building. This article maps the skills the Ops Atlas from xOps School associates with the discipline, shows how they connect, and proposes a ninety-day learning route.

Start with a candid observation from the Atlas: no standards body defines the term, so job titles vary widely. Some employers say threat intelligence analyst, others say threat hunter, and a few use ThreatOps as a team label. Read postings by their duties rather than their titles.

The two halves of the ThreatOps role

The Atlas takes the editorial position that intelligence and hunting belong in one operation. That implies a hybrid skill profile. The intelligence half is about question-setting, source evaluation and writing. The hunting half is about hypotheses, query writing and data fluency. Many people arrive from one side. Operations engineers usually bring data skills and need to build analytic writing; analysts usually bring judgement and need to deepen their comfort with logs.

Skill 1: writing questions that decision-makers can use

The entry advises keeping five to eight priority intelligence requirements active, each carrying a lone question with the single fact and single decision attached, reviewed quarterly. For a career, this is the skill that separates a note-taker from an adviser. Practise by taking a vague worry, such as “are we exposed to ransomware?”, and splitting it until each piece could be answered with a single observation. Keep a portfolio of before-and-after examples.

Skill 2: judging sources

Learn the FIRST-style grid in which reliability runs from A to F and credibility from 1 to 6, and practise assigning grades to real public reports. A brand-new feed from a trusted provider, according to the Atlas, starts at A3. The grid teaches a habit hiring managers value: separating who said something from whether the claim holds up.

Skill 3: calibrated language

Analysts are judged by how clearly they express uncertainty. The Atlas refers to the NATO estimative scale, where “likely” represents roughly sixty to ninety percent, and to a convention that a sentence takes a probability term or a confidence statement, but not the pair together. Write ten short assessments using the rule and ask a colleague to restate each one; if their restatement differs from your intent, rewrite.

Skill 4: structured analytical techniques

Two methods appear in the Atlas for disputed judgements: the key assumptions check and analysis of competing hypotheses. Both are cheap to learn, and demonstrating them in an interview shows that you can reason under disagreement. Use them on a published incident report and write up what changed in your conclusion.

Skill 5: reading and using attacker-behaviour frameworks

ATT&CK is the common vocabulary. The Atlas notes that version 20 arrived in October 2026 with fifteen Enterprise tactics and 227 techniques, adding Stealth and Defense Impairment as tactics and breaking older mappings. Learn to map a report to techniques, and also learn the limits: tactics are not ordered steps, and not every technique deserves a detection. An applicant who can explain why a coverage heat map is a vanity measure stands out.

Skill 6: exchange formats and handling labels

STIX 2.1 and TAXII 2.1, both standardised by OASIS in 2021, move structured information; the traffic-light scheme in its second version, authoritative since August 2022, governs who may see it. Build a small lab: create a few STIX objects with expiry dates, run them through an open platform such as MISP, and mark them with handling labels. The Atlas notes that the label scheme needs a STIX extension, a detail that makes a good interview answer.

Skill 7: hunt design

The PEAK approach cited by the Atlas structures a hunt around actor, behaviour, location and evidence, limits scope to one or two behaviours and expects roughly a fortnightly cadence. Hunts must end in an artefact, and a competent hunter can explain why a negative result with complete data is still a result. Keep a notebook of five hunts, each with a hypothesis, queries and a closing artefact. Add the rerun test: ask a peer to repeat one from your notes.

Skill 8: measuring your own work

The Atlas offers four outcome ratios, covering question coverage, intelligence that led to action, hunt conversion and gap closure, and warns against counting indicators ingested or reports published. Being able to present those ratios for a small personal project is rare and persuasive.

Soft skills that decide promotions

Technical ability gets you into the room; communication keeps you there. A ThreatOps practitioner writes short briefs for people who will never read a log line, negotiates with the team that owns alert rules, and explains to an executive why a quiet quarter can still be a good one. The Atlas hints at this when it describes strategic briefs in the ninety-day plan and the handoff rule that sends a validated hunt result to the detection owner as a request. Both depend on trust built through clear writing and reliable follow-through.

Practise by writing one-page summaries after every hunt and sharing them with a non-specialist reader. If they can say what decision you need from them, the summary works. If they cannot, trim it until they can.

Where this fits in an operations career

People who already work in operations will notice overlap with incident-driven practice. Our piece on sharpening incident management with analytics-driven operations covers the response-side habits that ThreatOps feeds, while a practical learning plan for DevOps, SRE and operations teams offers a template for sequencing new skills. ThreatOps does not replace those; the Atlas scopes live incident handling and the always-on alert desk outside it, so you will collaborate with colleagues in those roles rather than compete with them.

A ninety-day route for a working engineer

  1. Days 1 to 30: write five questions for a system you know well, build a source register with grades, and put expiry dates on a sample indicator set. Join a community where information is exchanged, as the Atlas advises.
  2. Days 31 to 60: run two hunts using the four-part hypothesis frame, and draft escalation and sharing rules. A good milestone is a first detection request that the owning team accepts.
  3. Days 61 to 90: compare a platform with a plain repository for storing your work, write a short strategic brief, and set up ratios. Decide, on evidence, whether a feed you use deserves renewal.

The maturity ladder, read as a career ladder

The Atlas ladder runs from absent, through consuming alert-only feeds, requirements-driven operation and a hunting programme, to an intelligence-led state. Employers sit at different rungs, and the rung shapes the job. At the consuming stage you will mostly curate and rate sources. At the requirements-driven stage you will manage the question register and a platform. At the hunting stage you will design and review hunts. Ask in interviews which rung the team occupies and what the exit condition to the next one is; the Atlas example is eighty percent or better coverage of your questions held across two successive quarters.

Choosing between building and joining

Another career question is whether to grow the capability inside one employer or to join a provider that does it for many clients. The decision tree in the Atlas is a handy lens. When a company has no owner for findings, it tends to buy managed hunting, so provider roles concentrate hunting skill. When a company has stakeholders hungry for intelligence and good telemetry, it builds, so in-house roles concentrate context and relationship skill. Neither route is better; they build different muscles, and moving between them after a few years is common.

Portfolio checklist

  • A question register with before-and-after rewrites.
  • A graded source list and a table of indicator lifetimes.
  • Five hunt notebooks, each ending in an artefact.
  • One page of outcome ratios from a self-run project.

For the broader context, including the timeline of frameworks and the decision tree for buying versus building, read the Atlas entry on ThreatOps before your next conversation about a role in this area.