> For the complete documentation index, see [llms.txt](https://deeptics.gitbook.io/deeptics-whitepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://deeptics.gitbook.io/deeptics-whitepaper/implementations/appendix-robotics-claude.md).

# Appendix "Robotics Claude"

## Appendix A — Prompt Template “Robotics Claude”

#### Planning

* **Goal**: \[state the robot/agent target]
* **Constraints**: \[physics, time, resources]
* **Success Metrics**: \[metrics and tolerances]
* **Seed & Randomization**: \[seed and domain randomization settings]

#### Tool-Use

* **Tools**: `scenario_builder`, `sensor_rig`, `controller_eval`, `metric_report`
* **I/O Contracts**: \[input/output schemas]
* **Checkpoints**: \[logging/checkpoint intervals]

#### Guardrails

* **Safety**: \[force/velocity limits]
* **Compliance**: \[asset licensing and privacy]
* **Reproducibility**: \[pin runtime and dependencies]

## Execution Template (YAML)

```yaml
plan:
  objective: "Evaluate warehouse robot path planner under dynamic obstacles"
  metrics: ["path_length", "collision_count", "time_to_goal"]
  constraints:
    max_speed: 1.2 m/s
    max_force: 40 N
  scenario:
    map: "warehouse_v2"
    obstacles: "dynamic_pallets"
    randomization:
      seed: 1337
      lighting: [low, medium]
  tools:
    - name: controller_eval
      params: { planner: "RRT*", horizon: 5s }
    - name: metric_report
      params: { export: ["csv","video"] }
guardrails:
  license_policy: "commercial_eval"
  runtime:
    physics_runtime: "deeptics-physx:1.0.0"
    sensor_pipeline: "deeptics-sensor:0.4.2"
```