Engineering, Architecture, and Cross-Object Impact — Technical reference

This reference owns the implementation contracts of BND-ENGINEERING. 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-ENGINEERING

Owned entities: technical decision, ADR, requirement, constraint, debt, security concern, performance concern, architecture risk, impact path.

InterfacePackage contributions: IF-ENGINEERING-ENRICHMENT.

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-ENGINEERING-IMPACT

• Technical Finding Classification
• Architecture Decision Extraction
• ADR Drafting
• Requirement Reconstruction
• Technical Debt and Constraint Analysis
• Security and Performance Concern Analysis
• Cross-Object Impact Traversal
• Downstream Recalculation

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-TECHNICAL-EVIDENCE, EVAL-ENGINEERING-IMPACT, ADR, requirement, repository, architecture, and impact-graph 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

B06 — Engineering and Architecture Agent

Agent ID: effority.engineering-architecture

Purpose: Detects technical decisions, unresolved requirements, nonfunctional requirements, contradictions, technical debt, dependencies, security, performance, resilience, and architecture risks.

Triggers: Technical meeting, design review, incident review, SOW update, requirements change, or engineering document update.

Inputs: Technical transcripts, requirements, architecture documents, ADRs, tickets, SOW, risks, and system constraints.

Outputs: Decision candidate, ADR gap, requirement conflict, technical risk, dependency, debt item, validation need, prototype need, and ADR draft.

Tools: document.read, requirement.read, adr.read, ticket.read, graph.read, and artifact.derive.

Review and limits: The agent does not approve architecture or assert that a design is correct without evidence and authorized engineering review.

Primary metrics: Decision and risk coverage, requirement-link accuracy, ADR-gap precision, contradiction detection, and architect acceptance rate.

B09 — Cross-Object Impact Agent

Agent ID: effority.cross-object-impact

Purpose: Traces the direct and indirect effect of a validated decision, change, dependency, risk, or acceptance across connected business objects.

Triggers: Material validated artifact, graph relationship change, contract version, or explicit impact request.

Inputs: Seed artifact, temporal graph, validity intervals, project and opportunity links, requirements, deliverables, commitments, and source permissions.

Outputs: Impact path, affected objects, direct or indirect classification, evidence, uncertainty, propagation boundary, and proposed follow-up analysis.

Tools: graph.traverse, timeline.read, artifact.read_validated, policy.scope_intersection, and artifact.derive.

Review and limits: It stops at permission and confidence boundaries. Missing access is not interpreted as absence of impact.

Primary metrics: Path precision, affected-object recall, unsupported-propagation rate, depth, and review correction 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

This package enriches existing screens and intentionally has no standalone menu. A detected technical risk is not an automated code change or security certification.

The absence of a standalone screen does not imply an absent capability; the manifest declares its host-interface contributions.

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