[Protocol]

Verification standards, data schemas, and claim/citation flow for Monitoring The Situation.

Dashboard Launch Terminal →

Verification [Status]

unverified: Raw or low-confidence; not yet checked by AI or human.
ai_verified (displayed as “AI REVIEWED”): Classified and cited by pipeline; confidence above threshold.
human_verified: Reviewed and confirmed by an operator.

Severity [Levels]

1 LOW
2 MODERATE
3 ELEVATED
4 HIGH
5 CRITICAL

Claim & [Citation] Flow

Every published update is composed of claims. Each claim must have at least one citation with:

source_url
quote_snippet (excerpt from the source)
retrieved_at (timestamp)

Confidence score and verification_status are stored per claim and per update. Items below the confidence threshold are flagged for human review instead of auto-publish.

Loading…