Answering with sources
Skill Scout-AI-Labs/scout-skills/skills/answering-with-sources
Agent Skills for using Scout: web research, sourced answers, company enrichment, structured extraction, monitoring. Works in Claude Code, Claude.ai, the Agent SDK, and Vercel eve.
npx -y skills add Scout-AI-Labs/scout-skills --skill answering-with-sourcesAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Answer a factual question quickly using the live web with Scout, with inline citations. Use for a single direct question that needs current information ("what changed in X", "who is Y", "is Z still true") when a full research pass would be overkill. One search, read the top results, answer with citations.
SKILL.md
1.1 KB, 212 tokens by cl100k_base, as published. Nobody here has run it
Answering with sources
The fast path: one search, read the best results, answer with citations. Use this for a single question; for a multi-part topic use researching-the-web instead.
Steps
- Search the question. Use
scout_search(MCP) or:
curl -s https://core.usescout.sh/v1/search \
-H "Authorization: Bearer $SCOUT_API_KEY" -H "Content-Type: application/json" \
-d '{"queries":["<question>"],"limit":8}'
- Read the top 2-3 results with
scout_scrape(orPOST /v1/page/markdown). - Answer in a few sentences, grounded only in what you read. Put a
[n]after each claim and list the source URLs at the end.
If the results don't actually answer the question, say that plainly and suggest a better query rather than padding.