# Project and Portfolio Execution — Technical reference

This reference owns the implementation contracts of `BND-PROJECT`. [Read the package guide](/documentation/packages/bnd-project) for usage, configuration, dependencies and upgrades. Generic execution, isolation and admission remain governed by [Platform Documentation](/documentation/platform).

## Data model and interfaces

**DataModelPackage:** `DM-PROJECT`

Owned entities: portfolio, project, workstream, plan snapshot, milestone, deliverable, progress, blocker, risk, issue, dependency, health vector.

**InterfacePackage contributions:** `IF-PROJECT-EXECUTION`.

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-PROJECT-DELIVERY`

- Project and Workstream Assignment
- Plan Comparison
- Progress and Blocker Analysis
- Risk and Issue Lifecycle
- Dependency Resolution
- Project Health
- Portfolio Aggregation
- Bottleneck Detection
- Capacity Concentration
- Intervention Prioritization

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-PROJECT-DISCLOSURE`, `EVAL-PROJECT-DELIVERY`, planning, issue-tracker, delivery, and capacity connector bindings.

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

### B02 — Project Execution Agent

**Agent ID:** `effority.project-execution`

**Purpose:** Detects progress, blockers, dependencies, risks, mitigations, schedule or cost deviation, quality concerns, and delivery-state changes.

**Triggers:** Project meeting completion, plan update, status report, ticket update, review outcome, or periodic project analysis.

**Inputs:** Validated meeting artifacts, plan, milestones, RAID, tickets, reports, project graph, and prior snapshots.

**Outputs:** Progress confirmation, blocker, dependency, deviation, risk change, mitigation, owner gap, project-health vector, and planning update draft.

**Tools:** `project.read`, `timeline.read`, `finding.read`, `evidence.search`, `ticket.read`, and `artifact.derive`.

**Review and limits:** Health remains a decomposable signal vector. The agent never declares contractual breach or customer acceptance as established fact.

**Primary metrics:** Finding precision and recall, project-link accuracy, blocker lead time, deviation-detection accuracy, and owner-gap coverage.

### B08 — Portfolio Delivery Agent

**Agent ID:** `effority.portfolio-delivery`

**Purpose:** Aggregates validated project evidence into cross-project dependencies, bottlenecks, capacity concentrations, customer risk, and intervention priorities.

**Triggers:** Periodic portfolio schedule, material project change, executive request, or resource-plan update.

**Inputs:** Authorized project health vectors, dependencies, milestones, risks, commitments, capacity information, customers, and delivery policies.

**Outputs:** Portfolio exception, dependency chain, concentration risk, delivery conflict, intervention priority, resource-action proposal, and portfolio narrative.

**Tools:** `portfolio.read`, `project.read_validated`, `graph.read`, `capacity.read`, `timeline.read`, and `artifact.derive`.

**Review and limits:** It does not rank employee value or allocate resources automatically. Proposed interventions remain decisions for authorized delivery leadership.

**Primary metrics:** Exception precision, dependency coverage, intervention acceptance, missed material risk, and evidence drill-down completeness.



## 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.

Delivery analysis consumes planning evidence and the B02/B08 outputs above. Contractual comparison and acceptance formalization are owned by [Scope Alignment](/documentation/packages/bnd-scope/technical-reference).


## Integration boundaries

A health indicator summarizes detected evidence, not ERP accounting. Missing or stale evidence remains visible and is not interpreted as healthy delivery.

Owned screen contracts: [SCR-PRJ-01](/documentation/packages/bnd-project/screens/intelligence-project-portfolio), [SCR-PRJ-02](/documentation/packages/bnd-project/screens/intelligence-project-projects), [SCR-PRJ-03](/documentation/packages/bnd-project/screens/intelligence-project-risks-and-issues), [SCR-PRJ-04](/documentation/packages/bnd-project/screens/intelligence-project-dependencies).

See [solution workflows](/documentation/technology-consulting-delivery/connected-workflows) for cross-package behavior. Screens, agents and workflows keep one canonical owner even when other bundles invoke or enrich them.