API access

One deterministic API behind every screen

Everything this site shows is a call to the same HTTP API: scenario engines, observed series, exports and the grounded AI reading. A partner's notebook, risk system or data warehouse can call it the same way.

https://api.videcra.live

Deterministic: The same inputs return the same run_id and the same numbers, today and after a redeploy. Runs are addressable and reproducible.
Versioned rulesets: Every response names its ruleset_version. Rules change only by a human edit, a version bump and a re-run of the golden scenarios.
Evidence on every channel: Each transmission channel carries its evidence grade — empirical, expert or hypothesized — its lag, and the source it rests on.
Never a forecast: Outputs are scenario illustrations conditional on the shock you specify. Nothing here is a prediction or investment advice.

Surface

Six endpoint families

Scenario engines

Rate transmission for the UAE, Saudi Arabia, the euro area and China; a US rate shock against a synthetic or your own uploaded book; chokepoints and natural catastrophe paths; compound briefings. The climate-transition contract is reserved and answers 501 until its source-pinned ruleset exists.

POST /scenarios/{uae,saudi,europe,china}-macro · /scenarios/rate-shock · /scenarios/custom-rate-shock (CSV) · /scenarios/chokepoint · /scenarios/natcat · /scenarios/climate-transition (reserved, 501) · /briefings/compound

Analytics

Observed anchors beside every modelled indicator, sensitivity grids, base / adverse / severe comparison, exposure contributions.

GET /analytics/{family}-anchors · /analytics/*-sensitivity · /analytics/uae-macro-mode-comparison · /analytics/rate-shock-contributions

Market context

Official series — FRED, ECB, IMF, BIS, World Bank and central-bank releases — with history for charting. Context for the reader, never an input to the engines.

GET /market-context

Exports

The extended scenario report as PDF or Markdown, with the observed-versus-implied tables, corporate layer, severity bookends and calibration truth points.

POST /export/pdf · /export/markdown · /export/report

Grounded AI reading

A brief or a structured analyst note written only from the run's facts. Every claim carries a run_id and a source key; a reply that cites anything else is rejected server-side.

POST /ai/explain {format: brief | note}

Workspace (signed in)

Snapshots, comparisons, briefs with review states and a change log, per account.

GET/POST /workspaces · /snapshots · /comparisons · /briefs · /change-records

Checking your session…

Access

Access for pilot partners

The scenario, analytics and market-context endpoints are open at pilot rate limits so the API can be evaluated without a key. Partner access is arranged by agreement and adds:

API keys per organisation, with usage visible to the organisation.
An audit trail — who ran what, when, against which ruleset version.
A stable, versioned contract: breaking changes announced ahead of time, never silent.
Your own exposures and parameters in a run, kept to your organisation.

Pilot limits today

Open endpoints are rate-limited per client, per API instance. Responses carry Cache-Control: no-store. No key is required for evaluation.

What the API guarantees