firecrawl-knowledge-base

Name: firecrawl-knowledge-base
Author: firecrawl

29kinstalls18forks

How to install firecrawl-knowledge-base?

firecrawl/firecrawl-workflows/firecrawl-knowledge-base

$npx -y skills add firecrawl/firecrawl-workflows --skill firecrawl-knowledge-base

Installs into the current project.

Summary

Build a knowledge base from web content with Firecrawl. Use for local reference docs, RAG-ready chunks, fine-tuning datasets, documentation mirrors, topic corpora, or LLM-ready markdown organized from web sources.

Files · 1

View on GitHub

SKILL.md

1	# Firecrawl Knowledge Base
2
3	Use this to turn URLs or topics into organized LLM-ready content.
4
5	## Onboarding Interview
6
7	Infer the source, goal, depth, and output location from context. If the source and goal are clear, proceed immediately.
8
9	Ask at most 1-3 concise questions only if blocked, such as the source URL/topic, whether the output is reference/RAG/training/docs, or training format if training is requested.
10
11	## Firecrawl Collection Plan
12
13	Use Firecrawl map for documentation sites, search for topic-based corpora, scrape pages into markdown, and preserve code examples and tables.
14
15	For files, follow the Firecrawl download-style convention:
16
17	```text
18	.firecrawl/
19	<hostname>/
20	<path>/
21	index.md
22	```
23
24	## Parallel Work
25
26	If appropriate, use sub-agents or equivalent parallel task runners:
27
28	- one docs section per researcher
29	- official docs, tutorials, community discussions, and references by source type
30	- source scraping vs chunk generation vs manifest generation
31
32	## Output Modes
33
34	- Reference: markdown files, `index.md`, and `sources.json`.
35	- RAG: markdown files plus chunk files and `manifest.json`.
36	- Training: scraped source files plus `training-data.jsonl` and `training-metadata.json`.
37	- Docs mirror: complete markdown mirror with a table of contents.
38
39	## Final Deliverable
40
41	```markdown
42	# Knowledge Base: [Source]
43
44	## Summary
45	[What was collected and why]
46
47	## Output Structure
48	[Files/directories created]
49
50	## Coverage
51	[Sections, source types, counts]
52
53	## Usage Notes
54	[How to use in RAG, docs, training, or agent context]
55
56	## Sources
57	[URLs collected]
58
59	## Rerun Inputs
60	workflow: firecrawl-knowledge-base
61	source: [url/topic]
62	goal: [reference/rag/train/docs]
63	depth: [quick/thorough/exhaustive]
64	output_dir: [.firecrawl/]
65	```
66
67	## Quality Bar
68
69	- Preserve code examples and formatting.
70	- Remove boilerplate navigation where possible.
71	- Include source URLs in frontmatter or metadata.

People who liked this also viewed

6 picks

Type

firecrawl-knowledge-base

29kinstalls18forks

View on GitHub

How to install firecrawl-knowledge-base?

firecrawl/firecrawl-workflows/firecrawl-knowledge-base

$npx -y skills add firecrawl/firecrawl-workflows --skill firecrawl-knowledge-base

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Firecrawl Knowledge Base
2
3	Use this to turn URLs or topics into organized LLM-ready content.
4
5	## Onboarding Interview
6
7	Infer the source, goal, depth, and output location from context. If the source and goal are clear, proceed immediately.
8
9	Ask at most 1-3 concise questions only if blocked, such as the source URL/topic, whether the output is reference/RAG/training/docs, or training format if training is requested.
10
11	## Firecrawl Collection Plan
12
13	Use Firecrawl map for documentation sites, search for topic-based corpora, scrape pages into markdown, and preserve code examples and tables.
14
15	For files, follow the Firecrawl download-style convention:
16
17	```text
18	.firecrawl/
19	<hostname>/
20	<path>/
21	index.md
22	```
23
24	## Parallel Work
25
26	If appropriate, use sub-agents or equivalent parallel task runners:
27
28	- one docs section per researcher
29	- official docs, tutorials, community discussions, and references by source type
30	- source scraping vs chunk generation vs manifest generation
31
32	## Output Modes
33
34	- Reference: markdown files, `index.md`, and `sources.json`.
35	- RAG: markdown files plus chunk files and `manifest.json`.
36	- Training: scraped source files plus `training-data.jsonl` and `training-metadata.json`.
37	- Docs mirror: complete markdown mirror with a table of contents.
38
39	## Final Deliverable
40
41	```markdown
42	# Knowledge Base: [Source]
43
44	## Summary
45	[What was collected and why]
46
47	## Output Structure
48	[Files/directories created]
49
50	## Coverage
51	[Sections, source types, counts]
52
53	## Usage Notes
54	[How to use in RAG, docs, training, or agent context]
55
56	## Sources
57	[URLs collected]
58
59	## Rerun Inputs
60	workflow: firecrawl-knowledge-base
61	source: [url/topic]
62	goal: [reference/rag/train/docs]
63	depth: [quick/thorough/exhaustive]
64	output_dir: [.firecrawl/]
65	```
66
67	## Quality Bar
68
69	- Preserve code examples and formatting.
70	- Remove boilerplate navigation where possible.
71	- Include source URLs in frontmatter or metadata.

People who liked this also viewed

6 picks

Type