A meditation timer with bells, an ambient loop and a session log is a sitting's work, and a model with a local text-to-speech voice can turn a script into audio with timed pauses. What Headspace sells is the library: guided sessions, sleepcasts, music and movement recorded by its own teachers and studios, plus a voice and a brand people already know. A synthetic voice reading an unreviewed script is a different experience, and free apps such as Medito and Insight Timer cover much of the same ground. Build the timer; decide on the subscription by the library.
Before you buildHeadspace shows a 14-day trial and no free plan on the pages we read. Medito is entirely free and Insight Timer has a large free tier, so try them before building or paying.
Build me a personal meditation timer and session player to use beside Headspace. Requirements: - Python 3.12 + FastAPI serving one plain HTML page with a little vanilla JS on http://127.0.0.1:4970, plus a terminal timer with `python -m sit timer 10`. - Timer: 3, 5, 10 and 20 minutes or a custom length; a start bell, an end bell and optional interval bells every N minutes, played from files in ./sounds (the README says to add your own). The page keeps the screen awake with the Wake Lock API and shows only the remaining time. - Ambient: loop a track from ./sounds with a fade in and out, or generate brown noise locally with `numpy`; the sleep timer fades anything playing to silence over the last 60 seconds. - Scripts: `python -m sit script "5-minute body scan"` asks an OpenAI-compatible endpoint (LLM_BASE_URL, LLM_MODEL, LLM_API_KEY in .env) for a calm script with lines like [pause 20s] and saves it to ./scripts/<slug>.md. Nothing is rendered until I have read the file. - Voice: `python -m sit render <slug>` reads each line with Piper (`piper-tts`, VOICE_FILE in .env), turns each pause into silence and joins everything with ffmpeg into ./data/audio/<slug>.mp3. - Log: sqlite3 at ./data/sit.db records every session (date, minutes, type). The home page shows a calendar grid, total minutes and a streak that forgives one missed day a week. - Single user. No accounts, no telemetry, local-first. The only outbound call is the optional `script` command. - Out of scope: recorded teacher content, courses, coaching, a mobile app and any claims about health or results. - README: where to find free bell sounds and Piper voices, and a plain note that generated scripts are unreviewed text to read first, and that this is a timer and player, not a course.
Opens your agent with the prompt filled in, you press enter. Or copy it and paste it into an empty folder.
The first prompt builds the core. Run these one at a time in the same folder to grow it into your own Headspace.
Add a sound mixer to the meditation timer in this folder. Requirements: - A Sounds page that lists the files in ./sounds/ambient and lets me play up to 4 at once with a volume slider each, using the Web Audio API with seamless looping. - Save mixes by name in ./data/mixes.json and add a sleep timer of 15, 30, 60 or 120 minutes with a 60-second fade out. - Keep the page working with the screen off: request a Wake Lock only while the page is visible. - Update the README.
Add a weekly review to the meditation timer in this folder. Requirements: - `python -m sit week` prints sessions, total minutes, the longest session and the days sat for the last 7 days, with the previous week beside it. - No scores or messages about results: counts and minutes only. - `--csv` writes the same data to stdout. Update the README.
Add a script library to the meditation timer in this folder. Requirements: - A Scripts page listing every file in ./scripts with title, length in minutes (from its pause lines and word count at 130 words per minute) and a Render button. - Rendered audio is cached by the SHA-256 of the script and the voice file name; editing the script re-renders it. - Edit scripts in a textarea on the page and show the rendered duration after saving. - Update the README.
People pay for a voice they already like and a library big enough that they never have to choose, with sleep audio, focus music and a daily session in the same app on their phone. The habit is helped by not having to find content each night.
Estimate · The timer, the log and Piper voices run locally for free. The optional script command costs a fraction of a cent per script through an API. The build is token usage on your agent plan.
Checked 2026-10-20 against Headspace's docs: 1, 2, 3, 4, 5, 6
| Plan | Monthly | Annual, per month | What you get |
|---|---|---|---|
| Individual | $12.99 | $5.83 | The full library of meditations, sleep sounds, yoga and fitness sessions and focus music, a new meditation each day and daily inspirational videos. |
| Family | — | $8.33 | The full library for up to six members, including the owner. |
| Student | — | $0.83 | The same library for verified college or university students aged 18 or over. |
| Mental health coaching | $99.99 | — | A separate subscription to sessions with a coach, not part of the app membership. |
New members get a 14-day free trial on the website; the pages we read show no permanent free plan.
Not really. People pay for a voice they already like and a library big enough that they never have to choose, with sleep audio, focus music and a daily session in the same app on their phone. The habit is helped by not having to find content each night. The prompt on this page builds the closest honest substitute.
About $12.99 a month on the Individual plan, $155.88 a year (checked 2026-10-20).
The library: guided meditations, sleepcasts, sleep sounds and focus music from Headspace's own teachers and studios; Move, the guided yoga and exercise sessions; A new meditation each day and the daily inspirational videos; Courses that run in order in one familiar voice; The phone and web apps, and the family plan for up to six people. If any of those matter to you, keep paying.
Yes: Medito (open source), Insight Timer (free), Smiling Mind (free).