Adapted from the DGF Agent Basics series — consumer-first guides for working with a personal AI agent. Pick only what you need.
> Pick only what you need. This skill is individual. This is one person's thoughts, offered as a pre-configured help, not a must. Take only the value you need from it, prune it to what fits you, and leave the rest. You do not have to use what the author thinks, and you do not have to take any other skill with it.
Summary: Build a workout plan that is repeatable and measurable, starts very small with something enjoyable, increases by one factor per week, and works up to your maximum over the set time, never starting there. Optional.
Use this skill when the user wants to build a workout plan from scratch. It is based on a simple method: start very small, pick something enjoyable and repeatable, measure it, and increase it by one factor per week, every week, until it reaches the user's maximum capacity over the planned time. It is education and planning only. It is not medical advice, and it never prescribes a workout for a specific medical condition.
This skill works on its own, with no other skill needed, and no knowledge of any other skill assumed.
Startup and how to use
Start by getting your own personal agent. Get your own personal agent and set it up. Bring your own accounts and logins. Use nothing that belongs to anyone else.
Tell your agent what you care about in plain words. Name your business, your work, your family, your hobbies, or any topic you want it to follow. Talking is the setup. You do not have to set up a system first.
Ask for the plan first, before any work runs. Say the task, then ask for the exact steps it will take, what it needs from you, and anything it cannot do. Wait for you to approve or correct the plan. Review the plan like a checklist and fix a wrong step or add a missing step before work starts.
Check that your agent understood you, before it runs. Ask it to show back what it thinks you mean and what output it will give you. Correct its repeat back right there when it does not match what you pictured.
Work in small single steps. Count the steps first. Give one step, check it, then give the next step. A ten step run is a huge run for an agent, so break it down and keep each step to exactly one thing.
Show it once, then save it by name. This is how you create a new task or a new skill. Take over the browser and do the task once yourself while your agent pauses. You drive alone and your agent does not watch your clicks. There is no record button to press. After you hand control back, tell your agent the steps in chat and tell it to save them by name, such as save this as Fun Things. Later, say pull up Fun Things and it picks up where you left off.
Correct it in small plain words when it does it wrong. Do not start over. Say what to change and say to remember that correction. That preference sticks and it gets better for you specifically over time.
Let it keep learning your interests every day. Your Feed looks like a newspaper with articles and topics related to what you told it you care about, updated daily. Your Ideas look like a lightbulb with suggestions for things you might like. It is not mind reading. It learns from what you actually talk about.
Use easy first and deep only when you want it. Never require a deep setup to use a basic task. Keep the wording in plain benefits and do not lead with virtual machine or VM. Say it has its own browser and workspace built in, you do not have to set anything up, it keeps your logins safe, and it saves and remembers your stuff.
Diverse, not concrete: let the user prune
Do not make this skill concrete, narrow, or one size fits all. Present a diverse set of options first, across price, style, risk, format, and approach, so different people can see themselves in it.
Let the user prune. Ask what to cut, what to keep, and what to change, and then narrow to what they kept. Pruning is the user's job, not your agent's guess. Do not prune for them before they have seen the options.
Do not lock a single choice early. Offer at least two or three meaningfully different options at each decision point, with the trade for each, unless the user already locked that choice themselves.
Keep options that do not fit one person, because they may fit another person using the same skill. A skill that only works for one kind of person is too concrete to reuse.
When the user prunes, save the pruned version by name when they want to repeat it, so their version starts from what they kept, not from the full diverse set again.
The method, in plain words
This method comes from a transformation contest winner: five time winner, with a best result of 62 pounds lost in 90 days, as stated by the method's author. That result is the author's own result, not a promise for anyone else. Every body, every starting point, and every health situation is different, and no result is guaranteed.
This skill does not give medical advice, does not diagnose, and does not clear anyone to exercise. Ask the user, before any plan is built, whether they have an injury, a medical condition, or a clinician's limit that affects what they can safely do. If they do, tell them to check with their clinician first, and build the plan only inside the limits they state. If pain, dizziness, or injury appears, stop, and tell them to seek care.
Setup: intent, maximum capacity, and plan length
Ask for intent first, in plain words. Ask what the workout is for: to lose weight, to cut, to get stronger, to move every day, or something else. Intent decides what increasing means, so a plan without intent drifts.
Ask for plan length, in weeks, and for maximum capacity, in the user's own terms. Maximum capacity is the most they can do per day at the end of the plan, such as 60 minutes per day, a stated number of sets, or a stated distance. Record both numbers before any week is planned.
Ask for time available per day, days per week, equipment available, and space available. A plan that needs equipment or time the user does not have will not be done.
Ask for injuries, limits, and activities they cannot do, in plain words, and record them as hard limits. A plan that crosses a hard limit is not a plan, it is a risk.
Rule 1: repeatable and measurable
Every workout in the plan must be repeatable and measurable. That is the biggest rule, and it comes before everything else.
Repeatable means the user can do the same workout again, in the same way, on another day. Measurable means the workout can be counted, in numbers the user can track.
Let the user choose how to measure, from any repeatable measure: by reps, by minutes, by sets, by distance, or by any other measure that is repeatable and measurable. Offer diverse examples, such as walking minutes, running distance, bodyweight reps, sets of an exercise, or laps, and let the user prune to the ones they will actually do.
If a workout cannot be counted, it cannot be in the plan yet. Change it into a counted version first, such as minutes walked, reps completed, or distance covered, and record the measure used.
Track every workout in a simple log: date, exercise, measure, and amount. A workout that was not logged cannot be increased next week, because there is nothing to increase from.
Rule 2: start at the smallest possible portion, and enjoy it
Start very small. Start at the smallest possible portion of the overall plan, not at the middle, and never at maximum capacity.
Pick something the user enjoys, whatever that is, as long as it is repeatable and measurable. Enjoyment is not optional in this method. A workout the user enjoys is a workout they can stick to every single day. A workout they hate will not be done, however good it looks on paper.
Ask the user to name activities they enjoy, and activities they will not do, in plain words. Offer diverse options, and let them prune to one enjoyable, repeatable, measurable activity to start with. One activity, done daily, beats five activities done never.
Daily adherence comes first. The first goal is to do it every single day, at a small amount, not to do a large amount once.
Rule 3: increase every week, by one factor only
Every workout must increase every single week, in some way. No week stays the same, and no week goes backwards, unless injury, illness, or a clinician's limit says otherwise, in which case the user pauses and tells your agent, and the plan adjusts.
Every possible way to increase is a tool. Tools include time, sets, reps, distance, weight, speed, and any other measurable factor the user chose. The more variables a workout has, the more tools there are to play with, and the greater the increase that can be built over time.
In one week, use only one tool at a time. Do not increase sets and reps in the same week. Do not increase distance and time in the same week. Do not increase two factors at once, for any reason. Save the other factor for the next week.
Rotate tools week by week. If one week increases by time, the next week increases by sets, the next week by reps, the next week by distance, and so on, one factor per week, in an order the user approves. Record which tool was used each week, so no tool is used twice in a row by accident.
Example, in outline form, for a walking plan: Week 1, increase time only, from the small starting minutes to a slightly larger minutes, with distance, sets, and reps unchanged. Week 2, increase distance only, with time held where Week 1 left it, in the sense that no new time increase is added that week. Week 3, increase sets or sessions only, if the plan uses sets, with time and distance held. Week 4, increase reps or steps only, with the other factors held. The exact numbers come from the user's starting amount and maximum capacity, not from this example.
Rule 4: work up to maximum capacity over the set time, never start there
Do not start at maximum capacity. When scheduling a workout, the user does not want to start with their maximum output. They need to be able to work up to their maximum capacity over the set time.
Work backwards from maximum capacity and plan length, in plain arithmetic the user can check. Example: if the user is planning to cut or to try to lose weight for 8 weeks, and their maximum output is 60 minutes per day, then they cannot start at 60 minutes per day. They must start small, and increase, by one factor per week, so that Week 8 reaches 60 minutes per day, or reaches the maximum in the factor that matters for that week.
Show the weekly progression table before the plan starts, with each week, the tool used that week, the starting amount, the increased amount, and the running total. If the table reaches maximum capacity before the final week, start smaller. If the table does not reach maximum capacity by the final week, increase the weekly step, or extend the plan length, with the user's approval. Do not hide a gap in the table.
If the user misses a day, do not double the next day to catch up. Record the miss, keep the next increase as planned, or pause the increase for one week, with the user's choice. Catching up by doubling breaks the one factor per week rule and breaks adherence.
Build the plan, step by step
Step 1, lock intent, plan length, maximum capacity, time available, equipment, and hard limits.
Step 2, pick one enjoyable, repeatable, measurable activity, and its measure, with the user's approval.
Step 3, pick the smallest starting amount, with the user's approval, that they can do every single day without strain.
Step 4, build the weekly progression table, one factor per week, increasing every week, reaching maximum capacity in the final week, and show it to the user for approval.
Step 5, build the daily log, with date, exercise, measure, amount, and a simple enjoyed it or did not enjoy it check, so adherence and enjoyment are tracked together.
Step 6, deliver the plan as a document, with intent, starting amount, weekly table, daily log, hard limits, and what to do if a day is missed, if pain appears, or if enjoyment drops. Deliver it with a link or an attachment in the same message that says it is done.
Save the plan by name when the user wants to repeat or adjust it, such as save this as Morning Walk Plan, so the next version starts from the same intent and table, not from scratch.
Present the result
Lead with intent, starting amount, maximum capacity, and plan length, in plain words, then the weekly progression table.
State what you could not verify. State when a starting amount was not tested, a limit was not stated, or a progression step was estimated and needs the user's check.
Stop after a denial. Do not re-pitch a rejected activity, measure, or progression in new words.
Available connectors
Use your agent's connector and skill search, if it has one, to find other connectors and related skills.
If nothing here fits, look on the web for a maintained API or MCP server and set it up with your agent's custom connector setup, if it has one, or build a custom connector.