# Text to Reminders
Turn free-form notes into an importable .ics reminders file with due dates, priorities and optional alarms — private, deterministic and in-browser.
## Run it
- **CLI:** `gizza tool text-to-reminders "Call the dentist tomorrow at 3pm
Urgent: submit expense report Friday
Buy milk"`
- **Web:** https://gizza.ai/tools/text-to-reminders/
- **Agents:** machine-readable descriptor (parameters JSON Schema) at https://gizza.ai/tools/text-to-reminders/tool.json
## Inputs
- `text` — Notes (one task per line) _(field)_
- `reference_date` — Reference date _(field)_
- `detect_priority` — Detect priority keywords _(field)_
- `include_undated` — Keep undated lines _(field)_
- `alarm_minutes` — Alarm minutes before due _(field)_
## Output
- ICS reminders (text)
## Query parameters
Open the tool pre-filled and auto-run via URL:
- `text` — Notes (one task per line)
- `reference_date` — Reference date
- `detect_priority` — Detect priority keywords
- `include_undated` — Keep undated lines
- `alarm_minutes` — Alarm minutes before due
Example: `https://gizza.ai/tools/text-to-reminders/?text=Call%20the%20dentist%20tomorrow%20at%203pm%0AUrgent%3A%20submit%20expense%20report%20Friday%0ABuy%20milk&reference_date=2026-03-02&detect_priority=true&include_undated=true&alarm_minutes=0`
---
## About this tool
**Text to Reminders** turns a pasted brain-dump into an importable iCalendar
`.ics` file. Write one task per line and the tool emits one reminder/task
component per non-blank line, with a due date when the line contains a recognised
phrase such as `tomorrow`, `Friday`, `in 3 days`, `2026-03-05`, `March 5`, or
`at 3pm`.
Parsing is deterministic: there is no LLM, no account, no upload and no guessing
beyond the documented date/time rules. Relative phrases are anchored on the
**Reference date** field so examples, CLI runs and page deep links are repeatable.
Priority keywords such as `urgent`, `asap`, `important` and `critical` can map to
iCalendar `PRIORITY:1`; low-priority words such as `someday` map to `PRIORITY:9`.
### Worked example
Input:
```text
Call the dentist tomorrow at 3pm
Urgent: submit expense report Friday
Buy milk
```
With reference date `2026-03-02`, priority detection on, undated lines kept and a
30-minute alarm, the output starts like:
```text
BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//gizza-ai//text-to-reminders//EN
CALSCALE:GREGORIAN
BEGIN:VTODO
UID:todo-1-20260302@text-to-reminders
DTSTAMP:20260302T000000Z
SUMMARY:Call the dentist
DUE:20260303T150000
BEGIN:VALARM
ACTION:DISPLAY
DESCRIPTION:Call the dentist
TRIGGER;RELATED=END:-PT30M
END:VALARM
END:VTODO
```
### Options
- **Reference date** anchors relative phrases (`today`, `tomorrow`, weekdays,
`in 2 weeks`). Leave it on today's date for real use, or set it explicitly for
reproducible CLI/page tests.
- **Detect priority keywords** maps `urgent`, `asap`, `important`, `critical` to
high priority and `someday`, `whenever`, `eventually` to low priority. Turn it
off to keep those words in the task title.
- **Keep undated lines** keeps lines with no recognised date as tasks without a
due date. Turn it off to emit only dated tasks.
- **Alarm minutes before due** adds a display alarm before each dated task. `0`
means no alarm.
### Limits & edge cases
- This is a deterministic parser, not an AI assistant. If a phrase is not in the
documented rule set, it remains part of the title instead of being guessed.
- Times are emitted as floating local iCalendar date-times (`DUE:YYYYMMDDTHHMMSS`)
with no timezone database. Your importing calendar app interprets them in its
local/calendar timezone.
- Repeating phrases such as `every Monday`, durations and end times are listed as
out-of-model in the analysis and are not expanded.
- Each non-blank line becomes one task. Multi-line task descriptions are not
merged; paste them as one line if they should stay together.
- iCalendar text values are escaped and long content lines are folded at the RFC
line length.
## FAQ
Does this use AI to interpret my notes?
No. The parser is a fixed Rust implementation that recognises a documented set of
date, time and priority phrases. That makes output reproducible and keeps the tool
fully local in the browser.
What date does "tomorrow" or "Friday" use?
Those phrases are measured from the **Reference date** field. A bare weekday means
the next occurrence after the reference date; for example, with reference date
`2026-03-02` (a Monday), `Friday` resolves to `2026-03-06` and `next Monday`
resolves to `2026-03-09`.
What happens to lines with no date?
When **Keep undated lines** is enabled, they are included as reminders/tasks with
no `DUE` property. When it is disabled, undated lines are skipped so the output
contains only reminders with a resolved due date or time.
Can I import the result into Apple Calendar, Google Calendar or Outlook?
Yes. The result is a normal iCalendar `.ics` document containing task/reminder
components. Calendar apps vary in how prominently they show imported tasks, but the
file is standards-shaped text and can be saved with an `.ics` extension.
## Related tools
- [Extract Action Items from Meeting Notes](https://gizza.ai/tools/action-item-extractor/): Extract action items, owners, and decisions from meeting notes or daily notes with deterministic rules. Markdown checklist or JSON, private in-browser.
- [Add Line Numbers](https://gizza.ai/tools/add-line-numbers/): Add line numbers to every line of text online, like nl or cat -n — custom start, step, separator, and alignment. Free and private, runs in your browser.
- [ANSI Log Renderer](https://gizza.ai/tools/ansi-log-renderer/): Paste ANSI-colored terminal output or CI logs and render them as HTML, or strip escape codes to plain text. Handles 16-color, 256-color, and truecolor SGR codes.
- [Rust AST Diff](https://gizza.ai/tools/ast-diff/): Compare two Rust source snippets structurally. The tool parses both files, canonicalizes their ASTs, and ignores formatting, whitespace, and comments.
- [Bulk Artifact Extractor](https://gizza.ai/tools/bulk-artifact-extractor/): Scan pasted text for emails, URLs, IPs, domains, phones, Bitcoin addresses, and Luhn-valid card numbers with offsets and context.