firecrawl-deep-research

Name: firecrawl-deep-research
Author: firecrawl

31kinstalls19forks

How to install firecrawl-deep-research?

firecrawl/firecrawl-workflows/firecrawl-deep-research

$npx -y skills add firecrawl/firecrawl-workflows --skill firecrawl-deep-research

Installs into the current project.

Summary

Produce an intensive, cited analytical report: executive summary, multi-angle findings, contrarian views, open questions, and full sources. Use only when the user needs rigorous synthesis of a complex topic (scientific, technical, policy, or market-analytical) that cannot be answ

Files · 1

View on GitHub

SKILL.md

1	# Firecrawl Deep Research
2
3	Use this only for report-scale research: a rigorous, cited synthesis the user
4	explicitly wants delivered as a formal written report. If the request is a
5	product pick, a top-N list, a quick lookup, or anything answerable with a short
6	search, stop; do not use this skill, let the request be handled the standard
7	way.
8
9	## Onboarding Interview
10
11	Infer the topic and output format from context. Before starting, unless already specified, always ask one short question to define the scope:
12
13	> "How long do you want this research task to run?"
14
15	Map the answer to a depth tier in the Collection Plan below:
16	- A few minutes → Quick
17	- ~10-15 minutes → Thorough
18	- Longer / no limit → Exhaustive
19
20	If the topic itself is unclear, you may ask at most 1-2 additional concise
21	questions (topic, or a critical angle/source constraint). Otherwise proceed once
22	the runtime is set.
23
24	## Firecrawl Collection Plan
25
26	Use Firecrawl search and scrape through the CLI or equivalent tool surface. Match
27	depth to the runtime the user chose during onboarding.
28
29	- Quick (~a few minutes): search 3-5 queries and scrape 5-10 high-quality sources.
30	- Thorough (~10-15 minutes): search 5-10 queries from different angles and scrape 15-25 sources.
31	- Exhaustive (longer): search 10+ queries and scrape 25+ sources, including primary sources, research papers, expert views, and contrarian sources.
32
33	Avoid re-scraping URLs already returned with full content from a search-with-scrape result.
34
35	## Parallel Work
36
37	If appropriate, use sub-agents or equivalent parallel task runners by research angle:
38
39	- overview and definitions
40	- technical or implementation details
41	- market and industry context
42	- contrarian views, risks, and limitations
43	- primary sources and official docs
44
45	Each researcher should return claims, source URLs, source quality notes, and uncertainty.
46
47	## Final Deliverable
48
49	Default structure:
50
51	```markdown
52	# Deep Research: [Topic]
53
54	## Executive Summary
55	[2-3 paragraphs]
56
57	## Key Findings
58	[Numbered findings with source links]
59
60	## Detailed Analysis
61	[Themes, evidence, and synthesis]
62
63	## Contrarian Views And Risks
64	[Counterarguments, limitations, failure modes]
65
66	## Open Questions
67	[What remains uncertain]
68
69	## Sources
70	[Every URL used with a one-line note]
71
72	## Rerun Inputs
73	workflow: firecrawl-deep-research
74	topic: [topic]
75	depth: [quick/thorough/exhaustive]
76	output: [markdown/json/brief]
77	```
78
79	## Quality Bar
80
81	- Cite sources for factual claims.
82	- Prefer primary sources when available.
83	- Flag uncertainty and conflicting evidence.
84	- Synthesize instead of listing scrape summaries.

People who liked this also viewed

6 picks

Type

firecrawl-deep-research

31kinstalls19forks

View on GitHub

How to install firecrawl-deep-research?

firecrawl/firecrawl-workflows/firecrawl-deep-research

$npx -y skills add firecrawl/firecrawl-workflows --skill firecrawl-deep-research

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Firecrawl Deep Research
2
3	Use this only for report-scale research: a rigorous, cited synthesis the user
4	explicitly wants delivered as a formal written report. If the request is a
5	product pick, a top-N list, a quick lookup, or anything answerable with a short
6	search, stop; do not use this skill, let the request be handled the standard
7	way.
8
9	## Onboarding Interview
10
11	Infer the topic and output format from context. Before starting, unless already specified, always ask one short question to define the scope:
12
13	> "How long do you want this research task to run?"
14
15	Map the answer to a depth tier in the Collection Plan below:
16	- A few minutes → Quick
17	- ~10-15 minutes → Thorough
18	- Longer / no limit → Exhaustive
19
20	If the topic itself is unclear, you may ask at most 1-2 additional concise
21	questions (topic, or a critical angle/source constraint). Otherwise proceed once
22	the runtime is set.
23
24	## Firecrawl Collection Plan
25
26	Use Firecrawl search and scrape through the CLI or equivalent tool surface. Match
27	depth to the runtime the user chose during onboarding.
28
29	- Quick (~a few minutes): search 3-5 queries and scrape 5-10 high-quality sources.
30	- Thorough (~10-15 minutes): search 5-10 queries from different angles and scrape 15-25 sources.
31	- Exhaustive (longer): search 10+ queries and scrape 25+ sources, including primary sources, research papers, expert views, and contrarian sources.
32
33	Avoid re-scraping URLs already returned with full content from a search-with-scrape result.
34
35	## Parallel Work
36
37	If appropriate, use sub-agents or equivalent parallel task runners by research angle:
38
39	- overview and definitions
40	- technical or implementation details
41	- market and industry context
42	- contrarian views, risks, and limitations
43	- primary sources and official docs
44
45	Each researcher should return claims, source URLs, source quality notes, and uncertainty.
46
47	## Final Deliverable
48
49	Default structure:
50
51	```markdown
52	# Deep Research: [Topic]
53
54	## Executive Summary
55	[2-3 paragraphs]
56
57	## Key Findings
58	[Numbered findings with source links]
59
60	## Detailed Analysis
61	[Themes, evidence, and synthesis]
62
63	## Contrarian Views And Risks
64	[Counterarguments, limitations, failure modes]
65
66	## Open Questions
67	[What remains uncertain]
68
69	## Sources
70	[Every URL used with a one-line note]
71
72	## Rerun Inputs
73	workflow: firecrawl-deep-research
74	topic: [topic]
75	depth: [quick/thorough/exhaustive]
76	output: [markdown/json/brief]
77	```
78
79	## Quality Bar
80
81	- Cite sources for factual claims.
82	- Prefer primary sources when available.
83	- Flag uncertainty and conflicting evidence.
84	- Synthesize instead of listing scrape summaries.

People who liked this also viewed

6 picks

Type