---

This section details on the original issue you should resolve

<issue_title>Improve error message when setpoint parameters are passed to follow_param</issue_title>
<issue_description>## Problem

When users include setpoint parameters (sweep parameters like B, T, Vtg, Vbg) in follow_param(), QCoDeS raises a cryptic error:

ValueError: Invalid dependencies

This error originates from QCoDeS' ParamSpecTree validation (dependency cycle detection) but gives no indication of what went wrong or how to fix it.

Feedback from AI agents (164 test runs)

This was one of the most common issues encountered (12 instances):

• "Initial sweep attempt failed from dependency cycles when following setpoint parameters"
• "ValueError: Invalid dependencies when follow_params included Vtg/Vbg (setpoint parameters)"
• "Resolved by following only measured signals (Lockin_Ix, Lockin_Vx)"
• "Initial init_database call failed (ValueError: ParamSpecTree cycles) when follow_param included setpoints"
• "Reran sweep without following the setpoint"

Root Cause

In base_sweep.py, _create_measurement() registers followed parameters with setpoints:

for p in self._params:
    if self.set_param is None:
        self.meas.register_parameter(p, setpoints=("time",))
    else:
        self.meas.register_parameter(p, setpoints=(self.set_param,))

When the user follows the set_param itself, QCoDeS detects a circular dependency (parameter depends on itself as a setpoint).

Proposed Solution

Add validation in follow_param() to catch this early with a clear error message:

def follow_param(self, *p):
    """Saves parameters to be tracked, for both saving and plotting data."""
    for param in p:
        # Check for dependency cycle: can't follow the setpoint parameter
        if self.set_param is not None and param is self.set_param:
            raise ValueError(
                f"Cannot follow setpoint parameter '{param.name}'. "
                f"The setpoint parameter is automatically recorded as the independent variable. "
                f"Only follow measured parameters (e.g., Lockin signals, voltmeters)."
            )
        self._params.append(param)

For Sweep2D, also check against both inner and outer set_params.

Benefits

• Clear, actionable error message instead of cryptic "Invalid dependencies"
• Catches the error early (at follow_param call) instead of at measurement start
• Educates users about the distinction between setpoints and measured parameters</issue_description>

Comments on the Issue (you are @copilot in this section)