More in Hobbies, Fitness & Learning.
Build a marathon training plan from 459 Strava runs with Claude
Beginner-friendlyNo logins, payments or coding needed. Low risk: paste the prompt and you should get something useful.
Copy the prompt below and paste it into the AI tool named on this page.
Prompt Improved prompt
Our fuller version of the source's prompt, or one written to avoid a problem in the source. Not tested by the poster. It doesn't change what happened in the source.
Before you start, ask me 1–3 short questions about the details below and wait for my answers. Once I've answered, confirm the plan in one line and run the whole task start to finish. Goal: Analyze my running history and build a marathon plan that fits my real weekly schedule, explained in plain language. Questions to ask me first: - When is the race, and what's your goal time? - Which days and times can you run each week? - What happened in your last race or training block? Inputs (fill in or confirm): - Running data export (Strava, Garmin or similar): [EXPORT] - Race date and goal: [RACE / GOAL] - Weekly availability: [SCHEDULE] - Injuries or limits: [INJURIES] Rules: - Explain what the data shows in plain words, not just charts. - Show how much of my running falls in each pace or heart-rate zone and what to change. - Build a week-by-week plan with recovery weeks, and say which research it's based on. - Leave room to swap or skip sessions when family life gets in the way. Checkpoint: Show me the analysis and plan outline before making a calendar file. If you get stuck: If my data is missing or unclear, ask instead of guessing, and tell me when to check with a coach or doctor. Verify before you finish: Check that weekly volume rises gradually and recovery weeks cut volume enough. Output: A plain-language analysis, a week-by-week plan and a calendar file with sessions on the right dates. Done when: I have a plan I understand that fits my week and is loaded into my calendar.
Customize this prompt
Answer these questions and we'll put your answers into the prompt for you. You can leave any of them empty.
Your prompt
Before you start, ask me 1–3 short questions about the details below and wait for my answers. Once I've answered, confirm the plan in one line and run the whole task start to finish. Goal: Analyze my running history and build a marathon plan that fits my real weekly schedule, explained in plain language. Questions to ask me first: - When is the race, and what's your goal time? - Which days and times can you run each week? - What happened in your last race or training block? Inputs (fill in or confirm): - Running data export (Strava, Garmin or similar): [EXPORT] - Race date and goal: [RACE / GOAL] - Weekly availability: [SCHEDULE] - Injuries or limits: [INJURIES] Rules: - Explain what the data shows in plain words, not just charts. - Show how much of my running falls in each pace or heart-rate zone and what to change. - Build a week-by-week plan with recovery weeks, and say which research it's based on. - Leave room to swap or skip sessions when family life gets in the way. Checkpoint: Show me the analysis and plan outline before making a calendar file. If you get stuck: If my data is missing or unclear, ask instead of guessing, and tell me when to check with a coach or doctor. Verify before you finish: Check that weekly volume rises gradually and recovery weeks cut volume enough. Output: A plain-language analysis, a week-by-week plan and a calendar file with sessions on the right dates. Done when: I have a plan I understand that fits my week and is loaded into my calendar.
Scroll to see the whole prompt. Your answers show in green.
4 blanks are still empty: Running data export (Strava, Garmin or similar); Race date and goal; Weekly availability; Injuries or limits. They'll stay in brackets in your prompt, so the chat tool may ask you about them.
What you type stays in this browser tab. We don't send it or save it, and it's gone when you leave the page. If you use a "Try this in" button, your answers go to that tool as part of the prompt.
Try this in:
This prompt has blanks to fill in, like [EXPORT]. We'll copy it instead of filling it in, so you can edit it before you send.
The source used Claude. These are general chat tools, so results may differ.
Opens in a new tab. We don't see what you send.
These links are only a shortcut. We're not affiliated with OpenAI, Google or xAI, and your results may not match the source.
The prompt they actually used
The exact prompt from the source.
Help me analyse my performance over the past year, and use that information to plan my training for the marathon.
AI tool
Outcome
After finishing a first marathon in 4:14 with a painful final stretch, Piet van den Boer asked Claude to analyze the race and training data. Claude found that 26% of the runs sat in a 'moderate' grey zone, too hard for easy days and too easy for quality days. For the next marathon, van den Boer gave Claude a full Strava export of 459 runs from three years, plus a weekly schedule built around two young kids. Claude built a specific, flexible plan, then reviewed it against marathon research and revised it with deeper recovery weeks (20–30% instead of 10–16%), truly easy Monday runs and strides on Fridays. Early on, Claude gave generic advice and a graph-filled PDF that van den Boer found useless, and a calendar export took several tries. Van den Boer says Claude could be confidently wrong and there's no way to tell without expertise.
Evidence: Described by posterThe person describes what happened in words, without screenshots or video.
Source
- Posted
- by Piet van den Boer (The long run, Substack)
- Last checked
- Method:
- How we verify
https://pvandenboer.substack.com/p/how-i-trained-for-a-marathon-with
Context: The post was written three weeks into the new plan, before the second marathon, so the result isn't known. Check with a doctor if you have an injury.
More in Hobbies, Fitness & Learning
Turn Claude into a blunt running coach with Garmin data and markdown files
Sander Sneekes's running had stalled: VO2max moved only from 44 to 45 in nine weeks, and the 10K time sat around 63 minutes for months. Sneekes connected Claude to Garmin data through an MCP connector and set up a project folder with three markdown files: coach instructions, a runner profile and a living analysis, plus one report per run. The rules told Claude to look at per-lap data and not to play along when Sneekes tried to call a mediocre run good. In the first week, an 'easy' run averaging 130 bpm, under the 135 target, turned out per lap to drift from 119 to 146, with nearly a third in Zone 3. A failed interval session led the coach to rewrite hard workouts into longer blocks or tempo runs on cool days, and it caught a too-fast cooldown, adding the rule 'cooldown means walking.'
PersonalHobbies & fitnessClaudeWorkedRecreated promptEvidence: Described by posterView CaseGet a feedback report after each Spanish practice chat with Langua
A Reddit poster used Langua for about a week, mostly for its AI chat and flashcards, to practice speaking Spanish. The poster found the feedback report after each chat 'super helpful': what went well, things to work on, specific corrections with explanations, then a cultural tip. Once, Langua offered a practice phrase in a tense that seemed wrong; when asked, it corrected itself and explained why. The poster called it worth the $20 a month subscription so far.
PersonalHobbies & fitnessLanguaWorkedRecreated promptEvidence: Described by posterBeginner-friendlyView CaseKnit doll socks from a ChatGPT pattern, then find one that works
Knitter Ciara of The Fairythorn asked ChatGPT to write a sock pattern for an 18-inch doll. Ciara managed to knit a sock from it, but only by turning the heel from experience, as that section had a fundamental error, and the sock came out far too small for the doll to ever wear. Asked instead to point to a free pattern on Ravelry, ChatGPT offered three suggestions in under a second. The first, 'Basic Doll Socks' by Jacquelyn Fitzgerald, knitted up quickly and easily and fit the doll, with just one small error.
PersonalHobbies & fitnessChatGPTPartialRecreated promptEvidence: Described by posterView CaseWrite a life story book with aphasia after a stroke, using ChatGPT to edit
A 75-year-old patient with aphasia from a stroke eight years earlier had trouble finding words and forming sentences, and wanted to write an autobiography. In speech therapy sessions, the patient wrote their own content and used ChatGPT to edit it, guided by a visual flow chart and set prompts. With ChatGPT, the patient produced more sentences with fewer errors. In three out of 10 sessions the patient noted ChatGPT sometimes wrote long, complicated sentences, and preferred the 'edit the following' prompt. It took clinician prompting and modeling at first. By semester's end there was enough for a printed life story book, shared with friends and family.
PersonalHobbies & fitnessChatGPTWorkedRecreated promptEvidence: Article or interviewView CaseHold a real Spanish conversation without translating in your head after a month with Langua
Christian Tapper has used the Langua AI tutor for over two years for Korean, Spanish and Finnish. A little over a month after starting Spanish on Langua, Tapper ran into an Argentinian man in a town in Finland and they talked in Spanish. 'For the first time, the words just came,' Tapper writes: no translating in the head, and expressions picked up in Langua chats. Afterward, a Mexican friend Tapper hadn't spoken to in nine months, other than by text, said the Spanish sounded much more fluent.
PersonalHobbies & fitnessLanguaWorkedRecreated promptEvidence: Described by posterBeginner-friendlyView CasePractice speaking Spanish before a move abroad with the Langua AI tutor
Spenser Mestel had studied Spanish on and off for two decades and was moving to Spain, so the goal was to get past a plateau. On Langua ($25 a month, about what Mestel pays for one virtual tutoring session), Mestel drew on work as a fitness coach and role-played a personal trainer meeting a new client, an AI character called Christian. Christian understood Mestel through the stumbles, and pausing to ask grammar questions brought 'some of the clearest explanations I've ever been given'. The limits: for accent, Mestel's biggest priority, it could only offer exercises, and Mestel kept meeting a human tutor, Maria.
PersonalHobbies & fitnessLanguaWorkedRecreated promptEvidence: Article or interviewBeginner-friendlyView Case
Related cases
Turn Claude into a blunt running coach with Garmin data and markdown files
Sander Sneekes's running had stalled: VO2max moved only from 44 to 45 in nine weeks, and the 10K time sat around 63 minutes for months. Sneekes connected Claude to Garmin data through an MCP connector and set up a project folder with three markdown files: coach instructions, a runner profile and a living analysis, plus one report per run. The rules told Claude to look at per-lap data and not to play along when Sneekes tried to call a mediocre run good. In the first week, an 'easy' run averaging 130 bpm, under the 135 target, turned out per lap to drift from 119 to 146, with nearly a third in Zone 3. A failed interval session led the coach to rewrite hard workouts into longer blocks or tempo runs on cool days, and it caught a too-fast cooldown, adding the rule 'cooldown means walking.'
PersonalHobbies & fitnessClaudeWorkedRecreated promptEvidence: Described by posterView CaseExplain English bills and prescription labels to parents in plain Spanish with Claude
Joel Salinas's parents don't speak English well, but almost everything that matters to them, such as a bill, a doctor's note or a prescription label, arrives in English. Now they take a photo and Claude translates it and explains it in Spanish, in plain terms they can act on, so a prescription becomes 'take one in the morning, one at night, with food.' Salinas calls it the use that reframed the others and says to verify the dose or the due date yourself.
PersonalCaregivingClaudeWorkedEvidence: Described by posterBeginner-friendlyView CasePrepare for a parent's cancer treatment talks with a tumor-board-style ChatGPT prompt
When John Bailey's mother's cancer came back, Bailey set up an AI assistant loaded with the mother's records and treatment history, using ChatGPT Projects and custom GPTs, to summarize test results and pathology reports in plain language. For big decisions, Bailey ran Deep Research in ChatGPT, Claude and Gemini with a prompt that had the AI reason as an oncologist, a radiation specialist, a molecular pathologist, a clinical-trials expert and a patient advocate, which produced a detailed, source-cited report. On at least two occasions it surfaced details that were initially overlooked, including a finding in a radiology report. The AI didn't choose treatments; it helped the family ask better questions.
PersonalCaregivingChatGPTClaudeGeminiWorkedRecreated promptEvidence: Article or interviewView Case