Build a Beneficial-Ownership Evidence Retention Review Console — implementation

## Build
Develop a beneficial-ownership evidence retention and permission review console using explicitly fictional policies. Parse BODS data to verify compliance with mock retention rules and generate structured permission evaluation reports.

## User
KYC onboarding escalation analyst

## Workflow
1. Parse simplified synthetic BODS v0.4 statements to extract beneficial ownership declarations and attached evidence document metadata.
2. Evaluate evidence records against explicitly fictional data retention rules and mock permission schemas.
3. Identify documents exceeding mock retention limits or lacking required user access level justifications.
4. Synthesize a structured permission review report detailing non-compliant records and recommended access revocation actions.

## Interface
Split-screen dashboard displaying evidence retention flags on the left and generated permission review reports on the right.

## Data
Generate simplified synthetic BODS statement packages in JSON format modeling multi-tiered holding companies and attached evidence metadata with fictional retention attributes.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Beneficial Ownership Data Standard](https://standard.openownership.org/en/latest/)
Extracts BODS v0.4 data modeling concepts for beneficial ownership declarations and entity relationship links.

## Tools
langgraph, pydantic, fastapi.
Manages complex state transitions across evidence parsing, mock policy evaluation, and report generation.
Fast hosted model or local Ollama instance capable of evaluating document metadata against structured mock policies and drafting clear reports.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Validates retention dates against fictional policy limits and identifies violations.
- Flags documents with unauthorized access according to mock permission schemas.
- Outputs a structured JSON report of retention violations and recommended actions.

## Starter Examples
```json
[
  {
    "input": {
      "subject_entity": "Atlas Logistics Holdings",
      "evidence_records": [
        {
          "doc_id": "doc1",
          "type": "passport",
          "retention_days": 400,
          "access_level": "admin"
        },
        {
          "doc_id": "doc2",
          "type": "utility_bill",
          "retention_days": 150,
          "access_level": "user"
        }
      ],
      "mock_policy": {
        "max_retention_days": 365,
        "required_access": "admin"
      }
    },
    "expected_output": {
      "subject_entity": "Atlas Logistics Holdings",
      "violations": [
        {
          "doc_id": "doc1",
          "reason": "Exceeds max_retention_days"
        },
        {
          "doc_id": "doc2",
          "reason": "Insufficient access_level"
        }
      ],
      "compliance_status": "failed",
      "recommended_actions": [
        "Revoke or delete doc1",
        "Update access for doc2"
      ]
    }
  },
  {
    "input": {
      "subject_entity": "Direct Green Energy SA",
      "evidence_records": [
        {
          "doc_id": "doc3",
          "type": "incorporation_certificate",
          "retention_days": 100,
          "access_level": "admin"
        }
      ],
      "mock_policy": {
        "max_retention_days": 365,
        "required_access": "admin"
      }
    },
    "expected_output": {
      "subject_entity": "Direct Green Energy SA",
      "violations": [],
      "compliance_status": "passed",
      "recommended_actions": []
    }
  }
]
```

## Safety
Performs read-only mock policy audits on simplified synthetic app records; uses explicitly fictional policies for compliance checks.
Use mock external writes. Require human review before any real deployment.