Firecrawl web research brief
Read-only Firecrawl research workflow. It searches web/news results through recued-core.firecrawl.web.search, renders result tables, and summarizes the returned titles, URLs, snippets, markdown, and warnings. It never starts crawl jobs, scrapes arbitrary routes directly, runs browser actions, sends custom headers/cookies, registers webhooks, or calls arbitrary Firecrawl API routes.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (13 steps)
search
?
web_results
default
Apply default
news_results
default
Apply default
image_results
default
Apply default
web_count
count
Count items in web results
news_count
count
Count items in news results
image_count
count
Count items in image results
summarize
?
web_rows
slice
Take a subset of
news_rows
slice
Take a subset of
web_table
to_table
Format results as a data table
news_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 Firecrawl search results for a product/research user: strongest facts, useful source URLs, gaps, disagreements, and next searches. Do not invent facts outside the Firecrawl response.
limit
setting
=
8
query
setting
=
formats
setting
=
markdown
sources
setting
=
[object Object],[object Object]
location
setting
=
firecrawl
setting
=
row limit
setting
=
12
max length
setting
=
900
time range
setting
=
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.