Skip to main content

A plain-text task manager and time tracker with pluggable AI parsing (Ollama / OpenAI-compatible / Anthropic).

Project description

jasem

A plain-text task manager and time tracker for your terminal, with pluggable AI parsing. One file, zero dependencies, your data in human-readable Markdown.

Install & run

Needs Python 3.8+.

pipx install jasem      # recommended — isolated, puts `jasem` on your PATH
# or
pip install jasem

Then write a task in plain language and see what's due:

$ jasem "pay rent next friday, high priority, finance"
✓ added #1: pay rent
  priority=high  deadline=2026-06-19  tags=finance

$ jasem today
Due today
  ☐   2  [medium]  2026-06-15 (today)  call dentist  #health

That's the whole loop: write naturally, jasem extracts the deadline, priority, and tags for you.

Features

  • Natural-language tasksjasem "submit report by friday, work" becomes a title, deadline, priority, and tags.
  • Views & editsjasem list, today, week, overdue, done 3, set 3 deadline tomorrow, rm 3.
  • Time trackingjasem track "1h 30min, code review, work", then jasem track week for per-day totals; fix or drop an entry with jasem track set 3 time 2h and jasem track rm 3.
  • Plain-text storage — everything lives in ~/.jasem/*.md; read, grep, edit, sync, or version-control it however you like.
  • Pluggable AI, local by default — runs fully offline with Ollama, or point it at any OpenAI-compatible API or Anthropic. If no model is reachable, tasks still save (dates parsed by regex).

Choosing an AI backend

Parsing is the only step that uses a model — pick one with environment variables:

# Local, free, private (default) — just have Ollama running
ollama serve && ollama pull qwen2.5:3b

# Any OpenAI-compatible API (OpenAI, Groq, OpenRouter, LM Studio, …)
export JASEM_PROVIDER=openai JASEM_API_KEY=sk-...
export JASEM_OPENAI_API_BASE=https://openrouter.ai/api/v1  # or OPENAI_BASE_URL

# Anthropic (Claude)
export JASEM_PROVIDER=anthropic JASEM_API_KEY=sk-ant-...

Persian (Jalali) calendar

Set JASEM_JALALI=true to work in the Persian (Jalali/Shamsi) calendar:

export JASEM_JALALI=true
jasem "pay rent 1405-04-10, finance"   # type deadlines in Jalali
jasem list                              # dates are shown in Jalali

Every date jasem shows is Jalali and every explicit date you type is read as Jalali — but your data on disk stays Gregorian ISO (YYYY-MM-DD), so the same files work with or without the flag. Relative phrases (tomorrow, next friday, in 3 days) work in both modes. Leave the variable unset (or false) for the default Gregorian calendar.

More

Run jasem help for the full command reference, every option, and all configuration variables.

License

MIT — see LICENSE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

jasem-0.7.0.tar.gz (39.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

jasem-0.7.0-py3-none-any.whl (40.1 kB view details)

Uploaded Python 3

File details

Details for the file jasem-0.7.0.tar.gz.

File metadata

  • Download URL: jasem-0.7.0.tar.gz
  • Upload date:
  • Size: 39.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for jasem-0.7.0.tar.gz
Algorithm Hash digest
SHA256 95bdefcebb21154a4d236dde32131327eb863ca4f74f5888756cae87cb210be7
MD5 bb3f26645d80c9d43d06a5a85072f208
BLAKE2b-256 8b2eb65f32e46329bdd46fbb1b9ba60e1517af19ddc43b8723b65b61846a0f71

See more details on using hashes here.

File details

Details for the file jasem-0.7.0-py3-none-any.whl.

File metadata

  • Download URL: jasem-0.7.0-py3-none-any.whl
  • Upload date:
  • Size: 40.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for jasem-0.7.0-py3-none-any.whl
Algorithm Hash digest
SHA256 9f30836dd9cac1b90f988770d04d6c082b25cdc79ec16b68bb87b8b9a69f7e62
MD5 be74fcf9f6900997cdd11c4a96c672cd
BLAKE2b-256 411ba1821341cc3b07672330a58aa6c5947bd26c8b0b84aee4dd6eac70391da5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page