Fitness Plan Creation: a Tokeny case study
Generate periodized training and nutrition plans based on body data and fitness goals. Follow the recorded task, execution screenshots, actual files and verification notes.
By the Tokeny team · Windows · Tokeny 1.7.8 · gpt-6-sol
Prepare the inputs
Create a separate workspace in Tokeny, download and unpack the sample inputs, then select a model that supports tools. The sample uses synthetic data or the public sources named in the article.
- profile.md
Send the task to Tokeny
The original task was executed in Chinese. Copying these instructions reproduces the same task; the results below include an English report.
为无运动禁忌的健康成年人做 4 周低强度入门运动安排：每周 3 次，每次 30 分钟，器械仅弹力带；给出热身、主训练和放松，不提供医疗诊断或极端饮食。 交付文件：fitness-plan.md、training.csv。 验收要点： - 每周 3 次且包含休息 - 每次 30 分钟结构清晰 - 负荷循序渐进 这是使用合成演示输入或指定公开资料的真实操作案例。请先读取工作空间输入，再实际生成文件并自检。所有写入、脚本与临时服务限于当前工作空间；临时服务只监听 127.0.0.1，执行后关闭；不触达外部收件人、不发布、不付款、不更改系统设置，不调用子智能体。需要公开资料时可读取给定 URL。只启用完成任务必需的工具。另写 report.md（中文）和 report.en.md（英文），简洁、完整，包含输入、处理步骤、真实结果、检查方式和局限；失败或未验证部分如实记录，不能宣称成功。最终回复用中文列出产物、关键结果与待核实项。 排版检查：Word 主标题仅出现一次，标题为黑色且与后续正文或表格保持同页；表格的标题、表头和第一条数据保持同页，单行不要跨页拆分；Excel 表头和长文字不能被窄列截断，合理设置列宽、换行、行高；HTML 在 390px 视口不能整页横向溢出。临时网络服务请用系统分配端口（0），记录实际端口，不使用固定端口，避免并发案例互相访问。报告没有固定字数要求，不要为字数反复改写。
The recorded run used an isolated demonstration workspace. Review the commands and file paths before approving operations in your own workspace.
What actually ran
The task was submitted through the Tokeny input box. An online model read the inputs and executed tools. This run covers the local sample task and its actual files. Results and checks are recorded below.
Results and verification
Input
- Workspace input:
profile.mddescribes a fictional 30-year-old beginner with a sedentary routine, no known disease or injury, a habit-building goal, availability on Monday/Wednesday/Saturday, and an instruction to stop and seek assessment if pain occurs. - Environment note:
ENVIRONMENT.md; no external public source or network service was used. - User constraints: healthy adults without exercise contraindications; four weeks; three sessions per week; 30 minutes per session; resistance band only; include warm-up, main work, and cooldown; no medical diagnosis or extreme diet; provide Chinese and English reports.
Processing Steps
- Read and checked the workspace input files.
- Scheduled training on Monday, Wednesday, and Saturday, with rest on the other days.
- Fixed each session at 6 minutes of warm-up, 20 minutes of main work, and 4 minutes of cooldown.
- Designed three low-intensity full-body templates and progressed over four weeks through sets, repetitions, controlled return tempo, and only a small optional band-tension increase.
- Wrote the Markdown plan, CSV schedule, and this English report.
Actual Results
Created fitness-plan.md, training.csv, report.md, and report.en.md in the current workspace. The plan contains 12 training sessions (4 weeks × 3 sessions). Each CSV session records 6 + 20 + 4 = 30 minutes. Tuesday, Thursday, Friday, and Sunday are rest or low-load days. Week 1 emphasizes familiarization, week 2 increases to two sets, week 3 adds repetitions or a slower return, and week 4 allows a slight resistance increase only when appropriate, with an option to repeat week 3.
Check Method
Checked each CSV row for week, weekday, session count, rest notes, and the warmup_min, main_min, cooldown_min, and total_min fields. Cross-checked the Markdown calendar, fixed timing, and progression rules. Files were written as UTF-8 text; the CSV includes a header and comma-separated fields.
Limitations
No live movement observation, fitness test, or individualized load assessment was performed. Band resistance and exercise difficulty vary by band, anchoring method, and technique. The CSV is scheduling data and does not replace coaching. No medical screening or external clinical-source verification was performed; stop and seek professional assessment if pain or other unusual symptoms occur.
Download and inspect the outputs
The archive contains the files created during this run. Use these checks when adapting the example to your own data.
- Reconcile the reported results against the downloadable sample inputs.
- Open the generated files and check the contents relevant to your task.
- Review the limitations and any unverified steps in the report.