Severity Assessments
Current and historical event-impact assessments, severity indexes, methodologies, evidence, uncertainty, and corrections.
Draft v0.1 — design contract, not yet a live service
This documentation set defines a proposed Time Seventeen protocol for review and implementation planning. https://api.t17.capital is the approved and reserved REST origin, but neither that origin nor any documented endpoint is claimed to be live. The WebSocket origin remains unspecified. Credentials, chain IDs, contract addresses, rate limits, and SLAs remain unapproved; IDs, prices, and timestamps shown throughout are illustrative, not real production data.
Severity Assessment measures forecast or observed real-world impact. It is independent from Market Rating and says nothing about market rules, liquidity, data quality, or tradability.
Endpoint summary
| Method & path | Purpose |
|---|---|
GET /events/{event_id}/severity | Current effective Severity Assessment |
GET /events/{event_id}/severity-history | Append-only assessment history |
GET /severity-indexes/{index_id}/value | Current derived Severity Index value |
GET /severity-indexes/{index_id}/values | Severity Index value history |
GET /severity-methodologies/{methodology_id}/versions/{version} | One immutable methodology version |
All endpoints are public.
T17 Severity scale
| Level | Meaning |
|---|---|
SEV-0 | No verified material impact |
SEV-1 | Limited/localized impact |
SEV-2 | Material but contained impact |
SEV-3 | Severe multi-system or regional impact |
SEV-4 | Extreme national/cross-border impact |
SEV-5 | Catastrophic/systemic impact |
UNKNOWN | Insufficient or conflicting evidence |
UNKNOWN is not equivalent to SEV-0. Missing evidence MUST NOT be interpreted as verified absence of impact.
Event assessment
{
"data": {
"assessment_id": "seva_demo_204",
"revision": 4,
"supersedes_id": "seva_demo_198",
"event_id": "evt_demo_001",
"assessment_type": "observed",
"level": "SEV-3",
"score": "67.5000",
"score_range": { "low": "61.0000", "high": "74.0000" },
"component_scores": {
"human": "58.0000",
"economic": "72.0000",
"infrastructure": "63.0000",
"geographic": "66.0000",
"geopolitical": "80.0000",
"duration": "66.0000"
},
"confidence": "0.7600",
"reason_codes": ["CROSS_REGION_EFFECT", "INFRASTRUCTURE_DISRUPTION"],
"rationale": "Illustrative concise impact rationale.",
"event_time_window": {
"start": "2026-08-01T00:00:00Z",
"end": null
},
"observation_cutoff_at": "2026-08-31T10:00:00Z",
"methodology_id": "sevm_demo_001",
"methodology_version": "1.1",
"issuer": {
"kind": "model",
"issuer_id": "model_demo_severity_001",
"model_version": "2026.08"
},
"evidence": [
{
"evidence_id": "evid_demo_sev_501",
"kind": "official_report",
"source_uri": "urn:t17:evidence:demo-impact-report",
"observed_at": "2026-08-31T09:30:00Z",
"content_hash": "sha256:ILLUSTRATIVE_ONLY"
}
],
"effective_from": "2026-08-31T10:00:00Z",
"effective_to": null,
"published_at": "2026-08-31T10:05:00Z",
"review_state": "approved",
"override": null,
"dispute": null,
"withdrawal": null,
"correction_reason": "Incorporated revised infrastructure evidence."
},
"meta": {
"request_id": "req_demo_severity_001",
"published_at": "2026-08-31T12:00:00Z",
"schema_version": "1.0"
}
}Forecast and observed timelines
assessment_type is forecast or observed. They are distinct series and MUST be filtered/labeled explicitly. A forecast carries forecast_horizon_end; an observed assessment carries an observation_cutoff_at. The current endpoint accepts assessment_type and MUST NOT merge the two into an unlabeled value.
Required dimensions
| Component | Measures |
|---|---|
human | Mortality, injury, displacement, health, and livelihood effects |
economic | Direct/indirect loss, output, trade, and fiscal effects |
infrastructure | Damage/disruption to critical physical and digital systems |
geographic | Spatial extent and population/economic area exposed |
geopolitical | Cross-border escalation, institutional, security, and diplomatic effects |
duration | Persistence, recovery horizon, and risk of recurrence |
Methodologies define ranges, weights, uncertainty treatment, evidence minimums, and level thresholds. Scores are fixed-point strings. Unknown components remain missing/unknown; they are not silently zeroed.
Assessment history
GET /events/{event_id}/severity-history accepts assessment_type, start_time, end_time, published_after, level, review_state, methodology_id, include_withdrawn, cursor, and limit. Default order is (effective_from, published_at, assessment_id) ascending.
Every revision preserves schema/methodology version, evidence IDs/hashes, issuer/model version, event time window, observation cutoff, uncertainty, effective interval, publication time, supersession, override, dispute, withdrawal, and correction data. Later evidence creates a new assessment; old records are never overwritten.
Review states are draft, under_review, approved, superseded, disputed, and withdrawn. Overrides and disputes follow the same append-only governance shape as ratings but operate under an independent severity methodology and authority.
Severity Index values
A Severity Index aggregates one or more Severity Assessments under its own immutable definition. It is not a tradable Market price and is not necessarily the same resource type as a prediction-market Index.
{
"data": {
"value_id": "seviv_demo_9001",
"revision": 1,
"supersedes_id": null,
"severity_index_id": "sevi_demo_global_001",
"severity_index_version_id": "sevivn_demo_003",
"value": "63.2000",
"unit": "severity_points",
"level_equivalent": "SEV-3",
"methodology_id": "sevm_demo_index_001",
"methodology_version": "1.0",
"input_assessment_ids": ["seva_demo_204", "seva_demo_199"],
"coverage_ratio": "0.9200",
"quality_state": "valid",
"calculated_at": "2026-08-31T10:10:00Z",
"source_cutoff_at": "2026-08-31T10:00:00Z",
"published_at": "2026-08-31T10:10:05Z",
"correction_reason": null
},
"meta": {
"request_id": "req_demo_sevi_001",
"published_at": "2026-08-31T12:00:00Z",
"schema_version": "1.0",
"max_age_ms": 600000
}
}GET /severity-indexes/{index_id}/values accepts the shared time range, cursor, and correction filters from Conventions. Each value retains all input assessment IDs so clients can reconstruct the calculation.
Methodology detail
GET /severity-methodologies/{methodology_id}/versions/{version} returns component definitions, ranges/weights, forecast and observed rules, uncertainty propagation, evidence hierarchy, minimum coverage, level thresholds, override/dispute governance, document URI/hash, publication time, and superseded version.
Not Market Rating
Severity responses never return MR1–MR5, liquidity, resolution clarity, or venue operational risk. See Market Ratings for those independent market-quality records.