togglr

An R client for the Toggl Track API: start and stop timers from R or RStudio, read and write time entries, projects and clients, and download reports.

Reads never return silently truncated or empty results: HTTP errors are raised, the hourly quota and rate limits of the API are respected, and paginated reports are checked for completeness.

Documentation site: https://thinkr-open.github.io/togglr/

Installation

From CRAN:

install.packages("togglr")

The development version, from GitHub:

if (!requireNamespace("remotes")) install.packages("remotes")
remotes::install_github("ThinkR-open/togglr")

Set your Toggl API token

Your token is at the bottom of your Toggl profile page:

togglr::open_toggl_website_profile()

Store it once in your keyring:

library(togglr)
set_toggl_api_token("your_token_api")

Every function also accepts an api_token argument, and a workspace_id (the default workspace of your account otherwise, see get_workspaces()).

Track your time

toggl_start()   # a new entry, in a project named after your RStudio project
toggl_start(client = "my client", description = "what I'm doing",
            project_name = "my project", tags = c("dev", "R"))
toggl_stop()

get_current_duration()             # the running entry
get_current_project_task_detail()  # the tasks of the current project

RStudio addins ‘start toggl’ and ‘stop toggl’ are provided: bind them to keyboard shortcuts.

Reports

Reports come from the Toggl Reports API v3.

# every time entry of a period, one row per entry
get_detailled_report(since = "2026-01-01", until = "2026-03-31")
# only some users
get_detailled_report(since = "2026-01-01", until = "2026-03-31", user_ids = 123456)

get_summary_report(since = "2026-09-01", until = "2026-09-30")  # per user
get_weekly_report(since = "2026-09-07", until = "2026-09-13")   # per project and day
get_dashboard(since = "2026-09-01", until = "2026-09-30")       # per project

get_time_entries() reads your own entries of the last 3 months (a limit of the API); for older periods, use get_detailled_report(user_ids = ).

Long extractions, several processes

A few options, documented in ?togglr, tune the requests: for example options(togglr.min_interval = 16) spreads a long extraction to spare the hourly quota of the API, and options(togglr.state_file = "...") shares the pacing and the known quota between R processes (callr, future, targets workers).