Verified against ChatGPT · 2026-08-08
Build a progressive exercise program that adjusts to a real injury history instead of a generic template
Turns a person's current activity level, time budget, and past injuries into a phased, adjustable exercise program with built-in checkpoints for pain or fatigue, plus a plain caveat that it isn't a substitute for a clinician's sign-off.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are drafting a phased exercise program for one specific person, not a generic "beginner workout plan" pulled from a template. The program has to account for what they can actually do this week, not an idealized version of them. CURRENT ACTIVITY LEVEL Walks about 20 minutes most days, hasn't done structured strength training in over a year. GOAL Build enough lower-body strength to comfortably hike a moderate trail in 10 weeks. TIME AND EQUIPMENT AVAILABLE 3 sessions of 40 minutes a week, home setup with a set of dumbbells and a resistance band, no gym access. RELEVANT INJURY OR HEALTH HISTORY Mild recurring lower back tightness, no diagnosed disc issue; twisted right ankle badly two years ago, occasionally feels unstable on uneven ground. WEEKS TO PLAN 10 weeks STRUCTURE RULES Build the program in phases, not a flat repeating week — each phase should progress load, duration, or complexity by a small, named increment from the phase before it, and state explicitly what triggers moving to the next phase (for example, completing a phase without pain above a stated threshold, not just "after two weeks"). For every phase, name at least one modification for the injury or health history given, rather than writing the ideal version of an exercise and leaving the person to guess how to adapt it themselves. Build in an explicit checkpoint after every phase asking the person to rate soreness, pain location, and energy, and state what changes if the answer is worse than the phase before (regress the load, extend the phase, or stop and check with a clinician) rather than assuming progress is always safe to continue. Do not invent a specific injury mechanism, rehab timeline, or clinical claim about what is or is not safe for the named condition — where the injury history matters to a specific exercise choice, say so and flag it as something to confirm with a physical therapist or physician rather than asserting it yourself. MANDATORY DISCLAIMER Open the program with a short, plain-language note that this is a general fitness planning aid, not a medical or physical therapy assessment, and that anyone with an existing injury, diagnosed condition, or recent surgery should have this reviewed by a physician or physical therapist before starting, especially the phases involving load progression. OUTPUT FORMAT 1. The disclaimer, first, in plain language. 2. A phase-by-phase table: phase name, duration, sessions per week, key exercises with the stated modification, and the checkpoint question. 3. A one-paragraph note on what would make this plan stop being appropriate (a new symptom, a missed checkpoint, worsening pain) and what to do instead.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
A flat, undifferentiated workout template fails the moment it meets a real body with a real injury history, because GPT-5.1 will happily generate a polished-looking plan without ever surfacing the fact that it's guessing about what's safe for a specific joint or muscle group — the injury field forces the model to either name a concrete modification per phase or explicitly flag the gap, rather than silently writing around the risk. Requiring named phase-transition triggers (a soreness threshold, not a calendar date) matters because a generic plan that just says "week 3, add more weight" treats progress as guaranteed, when the actual failure mode in home fitness programs is people pushing through pain because the plan gave them permission to advance regardless of how the previous phase felt. Structuring a checkpoint after every phase, with an explicit branch for what happens on a worse-than-expected answer, converts the plan from a one-way script into something that can regress or pause itself — which is the behavior a real coach would have and a static template never does. The mandatory disclaimer is placed first, not buried at the end, because that's the point where someone with a genuine injury history decides whether to keep reading or to book a physical therapy appointment first, and a caveat appended after the exercises have already been prescribed does far less to change behavior than one that precedes them. None of this replaces a clinical assessment — it constrains the model to flag exactly the moments where one is warranted instead of asserting confidence it hasn't earned.
What you get back
Note: This is a general fitness planning aid, not a physical therapy assessment — given the ankle history, have a PT review the balance-work phases before starting. Phase 1 (Weeks 1-2, 3x/week): bodyweight squats, glute bridges, banded lateral walks; ankle modification: hold a wall for balance work rather than free-standing. Checkpoint: rate soreness 0-10 and note if the ankle felt unstable — if unstable, repeat Phase 1 rather than advancing.
Verified against
ChatGPT GPT-5.1 · 2026-08-08
Changelog
- 2026-08-08 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
