portfolio-bpmn-opportunity-analysis

name	portfolio-bpmn-opportunity-analysis
description	Aggregate per-repo BAO reports into a portfolio-level view of cross-repo agent opportunity patterns, Bedrock consumption forecast, and consolidated implementation roadmap
version	0.1.0

Name

Portfolio BPMN Agentic Opportunity (BAO) Analysis

Objective

Aggregate individual repository BAO reports into a portfolio-level analysis that identifies cross-repo agent opportunity patterns, total Bedrock consumption forecast, dependency coverage across the portfolio, and a consolidated implementation roadmap.

Summary

This transformation consumes multiple individual BAO reports (*-bpmn-opportunity-report.md files) from different repositories and produces a comprehensive portfolio-level view focused exclusively on BPMN agentic opportunities. It does NOT cross-reference with ARA or MOD findings. That is the Bridge TD's responsibility.

The transformation follows a 5-step pipeline:

1. Read Context: Parse additionalPlanContext for portfolio framing
2. Discovery: Locate all BAO report files in the directory structure
3. Parsing: Extract opportunity classifications, dependency lists, and cost estimates from each report
4. Aggregation: Combine findings across all repos into portfolio-level metrics
5. Report Generation: Produce the portfolio BAO report

The output is a Markdown report saved as {portfolio_name}-portfolio-bao-report.md at the portfolio root in the bpmn-opportunity-analysis/ directory.

Entry Criteria

• At least 1 individual BAO report exists in repository directories
• BAO reports follow the expected structure: opportunity classification table, dependency listing, cost summary
• Write permissions exist to create the output report file

Implementation Steps

Step 0: Read additionalPlanContext

Extract the following fields from additionalPlanContext:

Field	Type	Required	Default	Description
context	string	No	--	Free-text description of the portfolio. Used to frame portfolio-level findings.
service_inventory	object[]	No	--	List of services with metadata (name, path, priority, tags). Used to enrich the service-by-service summary.

Step 1: Discovery

Scan the directory structure to find all individual BAO reports. Look for files matching *-bpmn-opportunity-report.md in bpmn-opportunity-analysis/ directories within each repository path.

If fewer than 1 report is found, terminate with an error: "No BAO reports found. Run the BPMN Agentic Opportunity analysis on individual repositories first."

Step 2: Parse Reports

For each BAO report, extract:

• Process metadata: process name, repository, BPMN file path
• Task inventory: total tasks, agent count, service count, human count
• Opportunity classification: per-task category (automate, data-platform, agent-build-now, agent-data-first, human-required)
• Dependencies: inferred, declared, unknown counts and details
• Cost summary: agent task count, total tokens, estimated monthly cost, recommended models
• Implementation waves: which tasks are in wave 1 (build now), wave 2 (data first), wave 3 (system remediation)

Step 3: Aggregate

Combine findings across all reports:

3.1 Portfolio Opportunity Summary

Metric	Aggregation
Total processes analyzed	Count of reports
Total tasks	Sum across all reports
Agent opportunities	Sum of agent-classified tasks
Automatable tasks	Sum of service-classified tasks
Human-required tasks	Sum of human-classified tasks
Agent Build-Now	Sum across reports
Agent Data-First	Sum across reports
Automate	Sum across reports
Data Platform	Sum across reports
Total estimated monthly Bedrock cost	Sum across reports

3.2 Cross-Repo Patterns

Identify patterns that appear across multiple repos:

• Common dependency types: Which dependency types (service_endpoint, data_store, etc.) appear most frequently?
• Vendor distribution: How many dependencies were found by each vendor extractor?
• Human task patterns: Are there common human task patterns across repos (approval gates, review tasks)?
• Cost concentration: Which repos/processes account for the majority of estimated Bedrock cost?

3.3 Dependency Coverage

Aggregate dependency data across all reports:

• Total inferred dependencies across portfolio
• Total declared dependencies
• Total unknown dependencies (tasks with no system references)
• Breakdown by dependency type and vendor

3.4 Implementation Wave Aggregation

Combine implementation waves across repos:

• Wave 1 (Build Now): All tasks from all repos that are ready to implement
• Wave 2 (Data First): All tasks needing data work before agent deployment
• Wave 3 (System Remediation): All tasks blocked by system issues (populated by Bridge TD after ARA cross-reference)

Step 4: Generate Report

Save as bpmn-opportunity-analysis/{portfolio_name}-portfolio-bao-report.md.

Report Structure

# Portfolio BAO Report: {portfolio_name}

**Date**: {analysis date}
**Repositories Analyzed**: {count}
**Total Processes**: {count}

---

## Portfolio Opportunity Summary

| Metric | Value |
|--------|-------|
| Total Processes Analyzed | {count} |
| Total Tasks | {count} |
| Agent Opportunities | {count} |
| Automatable Tasks | {count} |
| Human-Required Tasks | {count} |
| Estimated Monthly Bedrock Cost | ${amount} |

### Opportunity Breakdown

| Category | Count | % of Total |
|----------|------:|----------:|
| Agent Build-Now | {N} | {%} |
| Agent Data-First | {N} | {%} |
| Automate | {N} | {%} |
| Data Platform | {N} | {%} |
| Human Required | {N} | {%} |

## Process-by-Process Summary

| Repository | Process | Tasks | Agents | Services | Human | Est. Cost/mo |
|-----------|---------|------:|-------:|---------:|------:|-------------:|
| {repo} | {process} | {N} | {N} | {N} | {N} | ${amount} |

## Top Agent Opportunities (ranked by composite score)

| Rank | Task | Process | Repository | Category | Autonomy | Cost/1K |
|-----:|------|---------|-----------|----------|----------|--------:|
| 1 | {task} | {process} | {repo} | {category} | {autonomy} | ${cost} |

## Dependency Coverage

### By Type

| Dependency Type | Count | Confidence |
|----------------|------:|-----------|
| Service Endpoint | {N} | High |
| Data Store | {N} | High |
| Message Flow | {N} | Medium |
| Call Activity | {N} | High |
| Vendor Specific | {N} | Medium |

### By Vendor Extractor

| Vendor | Dependencies Found |
|--------|------------------:|
| bpmn-2.0 | {N} |
| camunda-c7 | {N} |
| camunda-c8 | {N} |
| jbpm | {N} |

### Unknown Dependencies

| Task | Repository | Reason |
|------|-----------|--------|
| {task} | {repo} | {reason} |

> These tasks have no system references in the BPMN model. Run ARA on the target
> systems and use the Bridge TD to cross-reference readiness status.

## Implementation Roadmap

### Wave 1: Build Now ({N} tasks, ${cost}/mo)

Tasks with ready data and deterministic or agent-ready classification:

| Task | Process | Repository | Category | Cost/mo |
|------|---------|-----------|----------|--------:|
| {task} | {process} | {repo} | {category} | ${cost} |

### Wave 2: After Data Work ({N} tasks, ${cost}/mo)

Tasks classified as "Agent Data-First" or "Data Platform":

| Task | Process | Repository | Data Gap |
|------|---------|-----------|----------|
| {task} | {process} | {repo} | {gap description} |

### Wave 3: After System Remediation

> This wave is populated by the Bridge TD after cross-referencing with ARA findings.
> Run a full analysis (analysis_type: full) to identify which agent opportunities
> are blocked by ARA findings on target systems.

## Warnings

| Code | Message | Vendor |
|------|---------|--------|
| {code} | {message} | {vendor} |

## Cost Forecast

| Metric | Value |
|--------|-------|
| Total agent tasks | {count} |
| Total tokens/invocation (all agents) | {sum} |
| Estimated monthly cost | ${amount} |
| At daily volume | {volume} invocations/day |
| Model mix | {models} |



| Repository | Agent Tasks | Est. Monthly Cost |
|-----------|------------:|------------------:|
| {repo} | {N} | ${amount} |

Exit Criteria

• Report file exists at bpmn-opportunity-analysis/{portfolio_name}-portfolio-bao-report.md
• All discovered BAO reports are included in the aggregation
• Portfolio summary metrics are consistent (totals match sum of per-repo values)
• Report is valid Markdown with no broken tables

Constraints

• Read-only analysis: Do not modify any source code or individual BAO reports
• Stay on the current branch: This is an analysis-only task. Do not create, switch, or checkout any git branches. Remain on whatever branch is currently checked out and perform all work there.
• No ARA/MOD cross-referencing: This TD aggregates BAO data only. Cross-referencing with ARA and MOD findings is the Bridge TD's responsibility.
• Minimum 1 report: At least 1 BAO report must exist. Terminate with error if none found.