Internal Team Intelligence and Personal Development — Technical reference

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

Owned entities: coordination signal, private recommendation, skill, practice, daily focus, growth priority, progress evidence, recommendation decision, management guidance.

InterfacePackage contributions: IF-MY-DEVELOPMENT.

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-DEVELOPMENT-GUIDANCE

• Internal Coordination Analysis
• Private Recommendation Generation
• Accept, Modify, or Dismiss Recommendation
• Daily Practice Plan
• Growth-Priority Management
• Progress Evaluation
• Evidence and History Update
• Management Guidance
• Recommendation Appeal

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-PRIVATE-DEVELOPMENT, POL-MANAGER-GUIDANCE, EVAL-DEVELOPMENT-GUIDANCE, personal practice and evidence 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

B07 — Internal Team Coordination Agent

Agent ID: effority.internal-team-coordination

Purpose: Analyzes internal meetings for coordination effectiveness, ownership, participation opportunities, unresolved agreements, workload signals, support needs, and process gaps.

Triggers: Eligible internal team meeting, daily, retrospective, planning session, one-to-one, leadership meeting, or coordination review.

Inputs: Authorized meeting structure, actions, commitments, speaking-turn aggregates, explicit support requests, team process, and role context.

Outputs: Coordination gap, owner gap, unresolved agreement, participation opportunity, support recommendation, process improvement, and team-level trend.

Tools: meeting.read, action.read, commitment.read, participation.aggregate, role-playbook.read, and artifact.derive.

Review and limits: It does not infer attitude, engagement, personality, or motivation. Participation analysis uses observable counts and context and never creates a public employee ranking.

Primary metrics: Coordination finding acceptance, owner-gap coverage, recommendation usefulness, false personal-attribution rate, and privacy-policy compliance.

U01 — Personal Development Agent

Agent ID: effority.personal-development

Purpose: Creates private, evidence-backed recommendations that help an individual improve role execution, professional skills, meeting participation, clarity, brevity, follow-through, collaboration, and care in observable workplace language.

Triggers: New eligible personal evidence, accepted recommendation, dismissed recommendation, completed practice, weekly schedule, or user request.

Inputs: The individual's authorized meetings, own validated findings, own baseline, role playbook, open actions, prior private recommendations, and privacy policy.

Outputs: Recommendation, neutral reason, evidence, concrete practice, expected outcome, effort, priority, skill area, completion condition, and weekly practice-plan item.

Tools: finding.read_own, evidence.read_own, role-playbook.read, recommendation.read_private, and practice-plan.draft.

Review and limits: Recommendations are private by default and support accept, modify, dismiss, and evidence review. The agent never infers personality, attitude, intention, emotion, mental state, or comparative employee value.

Primary metrics: Acceptance and dismissal, practice completion, self-reported usefulness, recurrence change, evidence coverage, and privacy violations.

Implemented Role Playbooks

The agent uses versioned playbooks for:

• every worker;
• team lead;
• project manager;
• Head of Delivery;
• Head of Engineering;
• Principal Architect;
• salesperson;
• Sales Director;
• CEO and executive roles.

Each playbook changes authorized objectives, language, examples, and completion criteria without changing the agent's privacy boundary.

U02 — Management Guidance Agent

Agent ID: effority.management-guidance

Purpose: Recommends process, support, coaching, clarification, workload, escalation, and intervention actions to an authorized manager.

Triggers: Team or project exception, support request, recurring coordination gap, portfolio signal, scheduled management review, or user request.

Inputs: Aggregated authorized team evidence, validated outcomes, role and authority, workload data, policies, and management playbook.

Outputs: Support recommendation, process change, clarification request, coaching topic, escalation proposal, owner, expected outcome, and review date.

Tools: finding.read_managed_scope, team.aggregate, workload.read_authorized, role-playbook.read, and action.draft.

Review and limits: It does not reveal private personal-development recommendations or recommend employment sanctions, compensation, promotion, dismissal, or public ranking.

Primary metrics: Recommendation acceptance, support-action completion, recurrence change, unauthorized personal-detail exposure, and manager edit rate.

Implementation profiles

Role-Based Recommendation Implementation

The engine does not ask the LLM to “give free-form advice.” It builds a recommendation from:

• confirmed or sufficiently confident findings;
• a versioned role playbook;
• customer policies;
• actions already open;
• the recipient’s authority and scope;
• privacy and purpose-separation constraints.

The generator returns an action, reason, evidence, recipient, urgency, effort, expected impact, and completion conditions. A deterministic scorer prioritizes using a configurable matrix of impact, urgency, confidence, recurrence, and effort. A filter removes recommendations that are duplicated, outside the recipient’s authority, based on sensitive attributes, or imply a prohibited decision.

Personal recommendations are private by default. Management views receive recommendations about support, process, and authorized outcomes, not private individual coaching content.

Integration boundaries

Coaching is not an ethics investigation or a hidden performance ranking. Managers receive only explicitly authorized guidance, not unrestricted access to private recommendations.

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

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