My relationships
Backed by the per-user MR precompute (one parquet per user). This is the individual counterpart to GET /v1/relationships: it returns only the people the bound user knows, sorted by that user’s own relationship_quality, and skips the team aggregation / overlay enrichment that the ring-wide endpoint does.
Requires a user-bound API key (USER-scoped, or a tenant key with the x-rings-user-id header) — returns 400 otherwise. Returns an empty list when the user has no relationships yet.
Authorizations
Query Parameters
Page number, 1-indexed. Clamped to >= 1.
Results per page. Clamped to the range 1-100 (values outside are silently capped, not rejected).
Field to sort by. Common values: 'relationship_quality', 'priority_score', 'last_activity_date' (default), 'company_score', 'person_name', 'company_name' — but any column name in the MRR parquet schema works (e.g. 'job_title', 'oldest_signal_created_at'). 'relationship_quality'/'priority_score' are aliases for the underlying 'highest_rq'/'highest_priority_score' columns. An unrecognized value silently falls back to 'last_activity_date' rather than erroring.
Sort order. Valid values: 'asc' (ascending) or 'desc' (descending).
asc, desc Comma-separated integers (1-3) to filter by exact match against each person's highest Relationship Quality (RQ) score across the ring.
Comma-separated integers (1-3) to filter by exact match against each person's highest Priority Score across the ring.
Filter by the person's primary company's AI score (company_score). Exclusive bound: matches scores strictly greater than this value (not >=).
Filter by the person's primary company's AI score (company_score). Exclusive bound: matches scores strictly less than this value (not <=).