Russia–NATO risk ladder

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Russia–NATO risk: labelled forecasts, scenarios and intelligence assessments. Cite russianato.site.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "Russia–NATO risk ladder": {
      "url": "https://mcp.russianato.site/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (8)

• get_overview read-only

  Start here. Annual risk for the five escalation levels today, a year ago and in one year (optimistic / baseline / pessimistic), plus average intelligence tone per country. Every number carries its provenance (published / model / manual).

  Parameters:

  • lang — string — Language of text fields (default en)

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_risk_ladder read-only

  Annual risk of each level on a date between one year ago and one year ahead of the data date. Past dates use the published forecast archive; future dates use the scenario model and also return cumulative risk from the data date. Optional what-if assumptions.

  Parameters:

  • date — string — YYYY-MM-DD; default = data date (as_of)
  • lang — string — Language of text fields (default en)
  • scenario — optimistic | baseline | pessimistic, or a number from -1 to 1
  • hybrid_pct — number — What-if: annual probability of a destructive hybrid attack, %
  • wide_share_pct — number — What-if: share of regional wars escalating into a wider war, %
  • scenario_factor — number — What-if: factor F by which intensity changes per year in the extreme scenarios
  • regional_share_pct — number — What-if: share of attacks escalating into a regional war, %

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_forecast_history read-only

  Published Metaculus probabilities by 2035 for an attack on European NATO or a seizure of Baltic territory, with the annual conversion for each observation.

  Parameters:

  • to — string — YYYY-MM-DD
  • from — string — YYYY-MM-DD
  • lang — string — Language of text fields (default en)
  • series — string

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_intelligence_assessments read-only

  Public assessments by intelligence services and governments of 12 countries, newest first, with hand-coded tone from -2 to +2 for a military attack and for hybrid attacks, averages by country, check dates and links.

  Parameters:

  • lang — string — Language of text fields (default en)
  • limit — integer
  • country — string — ISO code: EE LV LT PL FI SE NO DK NL GB DE US
  • dimension — string

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_sources read-only

  The 51 sources with what each says, how it is used, countries, check date and link. Filter by type (intel, research, method, forecast), country or text.

  Parameters:

  • lang — string — Language of text fields (default en)
  • type — string
  • limit — integer
  • query — string
  • country — string

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_other_forecasts read-only

  Prediction markets and other numeric forecasts with a different event or horizon (Manifold, Polymarket, Metaculus, Good Judgment Open, Atlantic Council), with check dates and corrections.

  Parameters:

  • lang — string — Language of text fields (default en)

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_changelog read-only

  Significant changes to data, model and site, and the corrections history (what was wrong, what changed, who reported it).

  Parameters:

  • lang — string — Language of text fields (default en)
  • type — string
  • limit — integer

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_methodology read-only

  Level definitions, formulas, current assumptions with check dates, the verbal probability scale and what the model deliberately does not do.

  Parameters:

  • lang — string — Language of text fields (default en)

  Hints: read-only yes · destructive no · idempotent yes · open-world no

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.