web_search

Name: web_search
Author: axoviq-ai

681stars61forks

How to install web_search?

axoviq-ai/synthadoc/web_search

$npx -y skills add axoviq-ai/synthadoc --skill web_search

Installs into the current project.

Summary

Search the web and ingest results as wiki pages

Files · 1

View on GitHub

SKILL.md

1	# Web Search Skill
2
3	Accepts a natural language query, calls the Tavily AI search API, and
4	returns the top matching URLs. Your agent receives those URLs and decides
5	what to do with them — fetch each one, display them, pass them to another
6	skill, etc.
7
8	## Setup
9
10	1. Install the dependency:
11	```bash
12	pip install tavily-python
13	```
14
15	2. Set your Tavily API key (free tier: 1,000 searches/month — sign up at
16	https://tavily.com, no credit card required):
17	```bash
18	# macOS / Linux
19	export TAVILY_API_KEY="tvly-your-key-here"
20
21	# Windows (Command Prompt)
22	set TAVILY_API_KEY=tvly-your-key-here
23
24	# Windows (PowerShell)
25	$env:TAVILY_API_KEY = "tvly-your-key-here"
26	```
27
28	3. Optional — cap the number of results (default: 20):
29	```bash
30	export SYNTHADOC_WEB_SEARCH_MAX_RESULTS=10
31	```
32
33	## Standalone usage
34
35	```python
36	import asyncio
37	from synthadoc.skills.web_search.scripts.main import WebSearchSkill
38
39	skill = WebSearchSkill()
40
41	async def main():
42	result = await skill.extract("search for: transformer architecture papers")
43	urls = result.metadata["child_sources"] # list[str] — top matching URLs
44	query = result.metadata["query"] # "transformer architecture papers"
45	print(f"Found {len(urls)} URLs for '{query}':")
46	for url in urls:
47	print(" ", url)
48
49	asyncio.run(main())
50	```
51
52	`result.text` is always empty — the skill is a discovery step that returns
53	URLs, not page content. Pass the URLs to the `url` or `youtube` skill (or
54	your own HTTP client) to fetch content.
55
56	## Intent prefixes
57
58	The skill strips a leading intent phrase before sending the query to Tavily:
59
60	\| Input \| Query sent to Tavily \|
61	\|---\|---\|
62	\| `search for: RAG evaluation` \| `RAG evaluation` \|
63	\| `find on the web: LLM benchmarks` \| `LLM benchmarks` \|
64	\| `look up quantum computing` \| `quantum computing` \|
65	\| `youtube: Karpathy transformers` \| `Karpathy transformers` (YouTube only) \|
66	\| `搜索: 深度学习架构` \| `深度学习架构` \|
67
68	YouTube-specific prefixes (`youtube:`, `search youtube:`, `youtube video:`,
69	etc.) restrict the Tavily search to `youtube.com` and `youtu.be`.
70
71	CJK intent phrases supported: 查找, 搜索, 网络搜索, 在网上查, 查一下
72
73	## Domain filtering
74
75	A built-in blocklist skips sites that block automated HTTP clients:
76	`reddit.com`, `medium.com`, `quora.com`, `twitter.com`/`x.com`,
77	`linkedin.com`, `wikipedia.org`, IEEE Xplore, ACM DL, and common
78	subscription-only academic publishers.
79
80	If `SYNTHADOC_WIKI_ROOT` is set, the skill also loads
81	`$SYNTHADOC_WIKI_ROOT/.synthadoc/blocked_domains.json` (a JSON array of
82	domain strings) to extend the blocklist at runtime.
83
84	## Scripts
85
86	- `scripts/main.py` — `WebSearchSkill`: intent parsing, domain filtering,
87	returns `child_sources` in metadata
88	- `scripts/fetcher.py` — thin async wrapper around `AsyncTavilyClient`
89
90	## Assets
91
92	- `assets/search-providers.json` — search provider registry (currently Tavily)
93
94	## Using with full Synthadoc
95
96	When running inside Synthadoc, the Orchestrator reads `child_sources` from
97	the result metadata and automatically enqueues each URL as a separate ingest
98	job, which are then processed by the `url` or `youtube` skill. No additional
99	setup is required beyond the env vars above.

People who liked this also viewed

6 picks

Type

azure-aiUse for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WHEN: AI Search, query search, vector searSkillsJul 2026472k1.3k
just-scrapeSearch, scrape, crawl, extract structured data, and monitor web pages via the ScrapeGraph AI CLI. Use when the user asks to search the web, scrape a webpage, grab content from a URL, extract JSON fromSkillsJul 2026245k37
firecrawlCLI and Agent Skill for Firecrawl - Add scrape, search, and browsing capabilities to your AI agentsSkillsJul 202694k531
firecrawl-searchWeb search with full page content extraction. Use this skill whenever the user asks to search the web, find articles, research a topic, look something up, find recent news, discover sources, or says "SkillsJul 202676k531
firecrawl-agentAI-powered autonomous data extraction that navigates complex sites and returns structured JSON. Use this skill when the user wants structured data from websites, needs to extract pricing tiers, producSkillsJul 202674k531
firecrawl-mapDiscover and list all URLs on a website, with optional search filtering. Use this skill when the user wants to find a specific page on a large site, list all URLs, see the site structure, find where sSkillsJul 202674k531

web_search

681stars61forks

View on GitHub

How to install web_search?

axoviq-ai/synthadoc/web_search

$npx -y skills add axoviq-ai/synthadoc --skill web_search

Installs into the current project.

Summary

Search the web and ingest results as wiki pages

Files · 1

View on GitHub

SKILL.md

1	# Web Search Skill
2
3	Accepts a natural language query, calls the Tavily AI search API, and
4	returns the top matching URLs. Your agent receives those URLs and decides
5	what to do with them — fetch each one, display them, pass them to another
6	skill, etc.
7
8	## Setup
9
10	1. Install the dependency:
11	```bash
12	pip install tavily-python
13	```
14
15	2. Set your Tavily API key (free tier: 1,000 searches/month — sign up at
16	https://tavily.com, no credit card required):
17	```bash
18	# macOS / Linux
19	export TAVILY_API_KEY="tvly-your-key-here"
20
21	# Windows (Command Prompt)
22	set TAVILY_API_KEY=tvly-your-key-here
23
24	# Windows (PowerShell)
25	$env:TAVILY_API_KEY = "tvly-your-key-here"
26	```
27
28	3. Optional — cap the number of results (default: 20):
29	```bash
30	export SYNTHADOC_WEB_SEARCH_MAX_RESULTS=10
31	```
32
33	## Standalone usage
34
35	```python
36	import asyncio
37	from synthadoc.skills.web_search.scripts.main import WebSearchSkill
38
39	skill = WebSearchSkill()
40
41	async def main():
42	result = await skill.extract("search for: transformer architecture papers")
43	urls = result.metadata["child_sources"] # list[str] — top matching URLs
44	query = result.metadata["query"] # "transformer architecture papers"
45	print(f"Found {len(urls)} URLs for '{query}':")
46	for url in urls:
47	print(" ", url)
48
49	asyncio.run(main())
50	```
51
52	`result.text` is always empty — the skill is a discovery step that returns
53	URLs, not page content. Pass the URLs to the `url` or `youtube` skill (or
54	your own HTTP client) to fetch content.
55
56	## Intent prefixes
57
58	The skill strips a leading intent phrase before sending the query to Tavily:
59
60	\| Input \| Query sent to Tavily \|
61	\|---\|---\|
62	\| `search for: RAG evaluation` \| `RAG evaluation` \|
63	\| `find on the web: LLM benchmarks` \| `LLM benchmarks` \|
64	\| `look up quantum computing` \| `quantum computing` \|
65	\| `youtube: Karpathy transformers` \| `Karpathy transformers` (YouTube only) \|
66	\| `搜索: 深度学习架构` \| `深度学习架构` \|
67
68	YouTube-specific prefixes (`youtube:`, `search youtube:`, `youtube video:`,
69	etc.) restrict the Tavily search to `youtube.com` and `youtu.be`.
70
71	CJK intent phrases supported: 查找, 搜索, 网络搜索, 在网上查, 查一下
72
73	## Domain filtering
74
75	A built-in blocklist skips sites that block automated HTTP clients:
76	`reddit.com`, `medium.com`, `quora.com`, `twitter.com`/`x.com`,
77	`linkedin.com`, `wikipedia.org`, IEEE Xplore, ACM DL, and common
78	subscription-only academic publishers.
79
80	If `SYNTHADOC_WIKI_ROOT` is set, the skill also loads
81	`$SYNTHADOC_WIKI_ROOT/.synthadoc/blocked_domains.json` (a JSON array of
82	domain strings) to extend the blocklist at runtime.
83
84	## Scripts
85
86	- `scripts/main.py` — `WebSearchSkill`: intent parsing, domain filtering,
87	returns `child_sources` in metadata
88	- `scripts/fetcher.py` — thin async wrapper around `AsyncTavilyClient`
89
90	## Assets
91
92	- `assets/search-providers.json` — search provider registry (currently Tavily)
93
94	## Using with full Synthadoc
95
96	When running inside Synthadoc, the Orchestrator reads `child_sources` from
97	the result metadata and automatically enqueues each URL as a separate ingest
98	job, which are then processed by the `url` or `youtube` skill. No additional
99	setup is required beyond the env vars above.

People who liked this also viewed

6 picks

Type