Build the application

## Build
Trace multi-tier corporate shareholding declarations to pinpoint unverified ownership thresholds. Synthesize formal documentary requests for missing intermediaries and produce unresolved control diagrams.

## User
Enhanced due diligence compliance officer

## Workflow
1. Ingest BODS v0.4 ownership and control statements linking corporate entities and holding vehicles.
2. Calculate aggregate direct and indirect share percentages to identify unexplained gaps above reporting thresholds.
3. Generate targeted Request for Information documents identifying specific missing intermediate entities.
4. Render structured graph representations distinguishing verified natural owners from unresolved holding nodes.

## Interface
Interactive graph inspector showing verified and broken control links, with an RFI document drafting modal.

## Data
Construct mock BODS v0.4 JSON datasets representing multi-tiered corporate holdings with deliberately omitted intermediary shareholdings.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Beneficial Ownership Data Standard](https://standard.openownership.org/en/latest/)
Applies BODS v0.4 data modeling standards to track entity-to-entity shareholding paths and calculate effective control.

## Tools
langgraph, networkx, pydantic.
LangGraph manages stateful traversal of multi-tier ownership chains and orchestrates evidence request drafting cycles.
Requires hosted model API for drafting customized regulatory information requests targeting missing beneficial owners.

## 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
- Detects breaks in ownership chains where aggregate declared control is below the statutory 25% threshold.
- Produces valid Mermaid or JSON graph representations highlighting unresolved corporate parent entities.
- Drafts itemized Requests for Information specifying exact legal entities and registration jurisdictions requiring documentation.

## Starter Examples
```json
[
  {
    "input": {
      "target_entity": "ENT-7788",
      "statements": [
        {
          "from": "ENT-7788",
          "to": "HOLDCO-1",
          "share": 0.6
        },
        {
          "from": "HOLDCO-1",
          "to": "TRUST-9",
          "share": 0.8
        }
      ],
      "natural_persons": []
    },
    "expected_output": {
      "unresolved_control_detected": true,
      "terminal_unresolved_entity": "TRUST-9",
      "effective_unresolved_share": 0.48,
      "evidence_request_items": [
        "TRUST-9 deed and trustee registry",
        "Beneficiary schedule for TRUST-9"
      ]
    }
  },
  {
    "input": {
      "target_entity": "ENT-1122",
      "statements": [
        {
          "from": "ENT-1122",
          "to": "PER-44",
          "share": 1
        }
      ],
      "natural_persons": [
        "PER-44"
      ]
    },
    "expected_output": {
      "unresolved_control_detected": false,
      "terminal_unresolved_entity": null,
      "effective_unresolved_share": 0,
      "evidence_request_items": []
    }
  }
]
```

## Safety
Operates entirely on synthetic corporate declarations without issuing binding legal notices or querying real registries.
Use mock external writes. Require human review before any real deployment.