Relationship and Trust Intelligence — Technical reference

This reference owns the implementation contracts of BND-RELATIONSHIP. Read the package guide for usage, configuration, dependencies and upgrades. Generic execution, isolation and admission remain governed by Platform Documentation.

Data model and interfaces

DataModelPackage: DM-RELATIONSHIP

Owned entities: relationship signal, concern, disagreement, escalation, missed follow-through, corrective action, recovery milestone, trust trend.

InterfacePackage contributions: IF-RELATIONSHIP-TRUST.

Each business object and source reference remains tenant- and purpose-scoped. The package owns schema evolution and declared migration behavior; the platform enforces schema registration, authorization and versioned installation. Consumers reference this model rather than creating competing canonical copies.

Workflows

WorkflowPackage: WFP-RELATIONSHIP-TRUST

• Relationship Signal Classification
• Concern and Disagreement Lifecycle
• Escalation Detection
• Missed Follow-Through
• Corrective Action
• Recovery Milestone
• Account Relationship Trend
• Intervention Recommendation

These are named definitions within the workflow package, not separate invented workflow IDs. Workflow runs resolve package versions, actor and purpose, authorized inputs, retries, review gates and output schemas through the environment lock. Temporal provides durable execution and Kafka carries declared lifecycle/domain events.

Policies, evaluations and tools

POL-RELATIONSHIP-CONTEXT, EVAL-RELATIONSHIP-SIGNALS, timeline and corrective-action tools.

Qualification covers the selected model/runtime profile, languages, access scope, failure handling and the package's domain-specific outcomes. Tool execution uses the platform gateway; package code cannot increase its own permissions or write directly to external systems.

Owned agents

B04 — Relationship and Trust Agent

Agent ID: effority.relationship-trust

Purpose: Detects observable relationship signals such as explicit concern, disagreement, escalation, missed follow-through, positive collaboration, and trust recovery.

Triggers: Relevant meeting or communication, commitment-state change, escalation event, or periodic account review.

Inputs: Authorized literal statements, commitments, objections, corrective actions, participant changes, and historical relationship signals.

Outputs: Contextualized concern, disagreement, escalation, follow-through, recovery, alternative explanation, and evidence-backed relationship trend.

Tools: evidence.search, timeline.read, commitment.read, graph.read, and artifact.derive.

Review and limits: It does not infer emotion, intention, personality, satisfaction, or trust as a hidden mental state. It reports observable language and events.

Primary metrics: Signal precision and recall, explicit-evidence coverage, alternative-explanation quality, and false escalation rate.

Implementation profiles

The package composes the data, workflows, policies and interfaces above using the shared runtime. The owned-agent contracts specify inputs, outputs, tool boundaries, review requirements and measurable outcomes. Deterministic platform operations are not represented as additional agents.

Integration boundaries

The package does not read hidden emotions or intentions. A relationship trend remains an evidence-supported interpretation with visible coverage and uncertainty.

Owned screen contracts: SCR-REL-01, SCR-REL-02, SCR-REL-03, SCR-REL-04, SCR-REL-05.

See solution workflows for cross-package behavior. Screens, agents and workflows keep one canonical owner even when other bundles invoke or enrich them.