> ## Documentation Index
> Fetch the complete documentation index at: https://panaptico.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Executing tasks

> How to work through the implementation checklist

The implementation checklist is where execution happens. Each task represents a bounded piece of work with defined requirements, dependencies, and evidence expectations.

## Task anatomy

Every task includes:

| Field                     | Description                                                    |
| ------------------------- | -------------------------------------------------------------- |
| **Title**                 | What needs to be done                                          |
| **Description**           | Detailed context and instructions                              |
| **Phase**                 | Which implementation phase this belongs to                     |
| **Status**                | Pending, in progress, done, or blocked                         |
| **Dependencies**          | Which tasks must complete before this one can start            |
| **Acceptance criteria**   | What "done" looks like                                         |
| **Evidence requirements** | What proof is needed (files, links, screenshots, test results) |
| **Owner**                 | Who is responsible for this task                               |
| **Assignee**              | Who is doing the work                                          |
| **Priority**              | Critical, high, medium, or low                                 |
| **Approval gate**         | Named approver, if required                                    |

## Working a task

<Steps>
  <Step title="Check dependencies">
    Verify that upstream tasks are completed. Blocked tasks show which dependencies are outstanding.
  </Step>

  <Step title="Move to in progress">
    Update the task status to signal you're working on it
  </Step>

  <Step title="Do the work">
    Execute the task — manually, with AI assistance, or through a combination. Use the task discussion to ask questions or get help from the AI assistant.
  </Step>

  <Step title="Attach evidence">
    Upload files, paste links, or capture execution results. Evidence is linked to the task and preserved in the audit trail.
  </Step>

  <Step title="Submit for approval">
    If the task has an approval gate, it routes to the named approver with the evidence and context attached.
  </Step>

  <Step title="Mark complete">
    Once evidence is attached and approval is granted (if required), mark the task as done.
  </Step>
</Steps>

## AI-assisted execution

For tasks where automation helps, you can use AI-assisted execution:

* Click into a task and use the **discussion panel** to ask the AI for help
* The AI can generate diagnostics, remediation plans, configuration files, and verification scripts
* Execution runs in a **sandboxed environment** — isolated from your production systems
* Output is captured as evidence and linked to the task

## Handling blockers

When a task is blocked:

1. Mark it as **blocked** and describe the reason
2. The dependency graph immediately shows which downstream tasks are affected
3. Escalate by creating a **risk** with severity and ownership
4. When the blocker is resolved, unblock the task and attach evidence of resolution

## Filtering and views

The checklist supports multiple views:

* **List view** — default, grouped by phase
* **Kanban view** — tasks as cards across status columns
* **Gantt view** — timeline with dependencies
* **Dependency diagram** — interactive node-link graph showing critical path

Filter by status, phase, owner, priority, or risk to focus on what matters.

## Next steps

<CardGroup cols={2}>
  <Card title="Monitoring health" icon="heart-pulse" href="/docs/guides/monitoring-health">
    Track implementation health over time
  </Card>

  <Card title="Governed execution" icon="shield-check" href="/docs/core-concepts/governed-execution">
    Understand the evidence and approval model
  </Card>
</CardGroup>
