Implementation brief

## Who Uses It
An operations team preparing new employees for their first day.

## What You Will Build
Build an agent that coordinates approvals, equipment, training, and account requests from a new-hire record. Give each team an owned task list and show blockers, completion evidence, and start-date changes.

## Inputs
- New-hire record
- Role-based onboarding checklist
- Task dependencies and team capacities

## Workflow
1. Create tasks from the approved onboarding checklist.
2. Assign owners and order dependent tasks.
3. Track updates from mock HR, IT, and delivery systems.
4. Replan incomplete work when dates or approvals change.

## Interface
An onboarding board with task owners, deadlines, dependencies, blockers, and a first-day readiness summary.

## Working Demo
Create the board, allow T2 and T4 to start, keep T3 blocked, and update readiness when the laptop is prepared.

## Deliverables
- A working local application with the specified review interface.
- Source code, dependency lockfile, and setup commands.
- A recorded end-to-end demo using the supplied starter data.
- Example outputs, action logs, and automated tests.

## Starter Data
These fictional records are a small development fixture, not a production dataset. Implement the mock services described in the brief.

```json
{
  "hire": {
    "id": "H1",
    "role": "support",
    "start_date": "2026-10-01"
  },
  "tasks": [
    {
      "id": "T1",
      "name": "Manager approval",
      "owner": "Manager",
      "depends_on": [],
      "status": "complete"
    },
    {
      "id": "T2",
      "name": "Prepare laptop",
      "owner": "IT",
      "depends_on": [
        "T1"
      ],
      "status": "pending"
    },
    {
      "id": "T3",
      "name": "Dispatch laptop",
      "owner": "Logistics",
      "depends_on": [
        "T2"
      ],
      "status": "pending"
    },
    {
      "id": "T4",
      "name": "Schedule training",
      "owner": "People Ops",
      "depends_on": [
        "T1"
      ],
      "status": "pending"
    }
  ]
}
```

## Constraints
Use fictional employees and mock systems. Do not ship equipment, sign documents, or create real accounts.
Choose an available model that supports your implementation. Record its identifier and version.

## Evaluation
Review the complete application, not only the starter-case answers.
- Every task has an owner and a dependency state.
- T3 cannot complete before T2.
- Readiness identifies incomplete tasks rather than claiming the hire is ready.
Add one normal case, one incomplete-input case, and one failure-recovery case.