Having written up a few such sets of course lab notes, the best advice I can give is to offer an "idiot-proof" introduction so that anyone can get the first lab started. From there, I tend to curve the difficulty of the tasks, such that the first task should be trivial, yet the last task of the session is more challenging. I then tend to suggest a couple of improvements or suggestions for the curious to consider or try out.
For subsequent labs, I presume knowledge from previous sessions is retained, and consider it a prerequisite. I give progressively less guidance on using tools or techniques for repetitive tasks through the different sessions.
To give an example, if it was a programming exercise, the first set of notes will be thorough and tested on an unfamiliar colleague, to ensure they can run the software needed, and get a "hello world" to run. This includes showing which buttons to press, and how to save and load work. While experienced students may skip this part, that's their choice... I find it worthwhile to write this, even if just to give the TA something to point students towards if they have problems.
As a student works through a given session in the notes, I make the tasks gradually reduce the direction given. So the first exercise could be to ask the user for their name and echo it back - I'd tell them to use the input command, then the print command. By the end of that session, the recap exercise night be to write a program to ask for 2 numbers and multiple them, returning the result. There wouldn't be any more guidance than that. For "optional extra ideas", I might suggest to keep asking for input until the user provides none, or something similar as a little teaser requiring some reading ahead and independent research.
The next session would not go into the same detail on getting set up, and would begin from a higher difficulty level (wouldn't step the student through which function to use, in the case of my trivial example above).
By the final session, there would be limited guidance and much more explanation of concepts or suggestion of references to look at. I favour leading people to good sources and suggesting they read it, rather than trying to reinvent the wheel, but this might not work in all cases.
In particular, physical laboratory activities with chemicals or equipment may not work with this model, as you may have obligations to detail all procedures for health and safety purposes. In that case, I would leave the analysis and explanation to the students, giving progressively less help in finding the right places to read about the expected results etc.
Ultimately this depends on the discipline and the expectations of the principal academic teaching the class. I do find students like the rising difficulty, as there's both sufficient help to get started, bus also a good sense of accomplishment to finish the exercises and have worked it all out.