How does the skill handle confidence and likelihood?

It applies a defined confidence scale to each judgement and a standardised likelihood yardstick to forward-looking statements, so terms like high confidence or likely carry a consistent meaning across every assessment you produce.

Is there a threat assessment template I can reuse?

Yes. The output follows a repeatable structure with a bottom line up front, key judgements, supporting analysis, likelihood-calibrated forecasts and intelligence gaps, and it is plain Markdown you can adapt to your house style.

What does it need to produce a good assessment?

It works from the evidence and sourcing you provide. The stronger your inputs, profiles, pivots and reporting, the better the judgements. It runs in Claude Code, Cursor, Codex or Windsurf and is free and MIT-licensed.

Threat Assessment.

/threat-assessment

A threat assessment is where evidence becomes a judgement a stakeholder can act on. This free, MIT-licensed skill runs inside your AI coding agent (Claude Code, Cursor, Codex or Windsurf) and helps you produce a structured assessment for an actor, sector or scenario, with a clear bottom line, analytic judgements separated from raw fact, calibrated confidence and standardised likelihood language throughout.

Get the free pack on GitHub →Or join the free tier waitlist

What it does.

The skill structures a threat assessment from the evidence you bring: a bottom line up front, the key judgements that follow from it, the reasoning and sources behind each, and the gaps or assumptions that qualify them. It keeps analytic judgements distinct from observed fact, attaches a confidence level to each, and applies consistent likelihood language so forward-looking statements mean the same thing every time.

When to use it.

Use it when a stakeholder asks how a given actor or scenario affects them, when a profile or campaign cluster needs turning into a judgement, or when you are producing a recurring sector assessment and want a consistent house format. It sits at the production end of the workflow, after profiling, pivoting and enrichment have done their work.

What you get back.

A finished assessment in a consistent structure: bottom line up front, key judgements with confidence levels, supporting analysis with sourcing, likelihood-calibrated forecasts, and a statement of intelligence gaps. The language is disciplined, so a reader can tell what you assess with high confidence from what remains uncertain, and act accordingly.

How it fits your workflow.

Run it from your AI coding agent and feed it the profiles, pivots and PIRs you have already built; the assessment ties them into a judgement and can be tagged against the requirement it satisfies. The Markdown output drops into a report or briefing, and the confidence and likelihood conventions match the rest of the rigour skills in the pack.

Keep going.

Frequently Asked Questions.

Want to see this on your own organisation?

Request a demo or start your free trial today, and get straight to AI-powered threat intelligence built around your organisation.