Hugging Face Daily Papers digest
Scheduled read-only Hugging Face research digest. It reads Daily Papers, recent Paper Pages, and documentation search context through recued-core.huggingface, summarizes notable AI research items, and can send an in-app notification when papers are visible. It never comments on papers, claims papers, mutates links, creates collections or discussions, downloads files, or calls arbitrary Hugging Face routes.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (17 steps)
daily_papers_raw
?
recent_papers_raw
?
paper_docs
?
daily_papers
default
Apply default
recent_papers
default
Apply default
daily_paper_count
count
Count items in daily papers
recent_paper_count
count
Count items in recent papers
has_papers
compare
Check if daily paper count is greater than 0
notify_enabled
compare
Check if setting notify when papers equals true
should_notify
all
Check if all conditions are true
summarize
?
notify
?
skip: step.should_notify not_equal true
daily_rows
slice
Take a subset of
recent_rows
slice
Take a subset of
daily_table
to_table
Format results as a data table
recent_table
to_table
Format results as a data table
card
to_summary
Format results as a summary card
Settings
Configurable at install. Defaults shown — change them anytime in Recued.
focus
setting
=
Summarize the most relevant Daily Papers and recent Paper Pages. Prioritize practical model, dataset, Space, benchmark, and implementation signals visible in the Hugging Face data. Do not invent facts outside the API responses.
limit
setting
=
10
channels
setting
=
in_app
end hour
setting
=
10
weekdays
setting
=
1,2,3,4,5
row limit
setting
=
10
max length
setting
=
1000
start hour
setting
=
8
huggingface
setting
=
notify when papers
setting
=
true
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.