oyi77 Overview
gate-news-briefing runs a news + sentiment briefing playbook using web search and public crypto-news sources. It assembles recent major events, trending stories, and a social-sentiment read. Standalone, tool-agnostic rewrite of the upstream gate-news-briefing skill (which used 3 MCP tools: news events, news feed search, social sentiment). Here those map to general web_search + source aggregation.
When to Use
- "What happened recently in crypto"
- "Today's crypto highlights"
- "Any new updates / news"
- "Crypto news briefing"
When NOT to Use
- A specific coin's fundamentals — use
gate-info-coinanalysis. - Market-wide price snapshot — use
gate-info-marketoverview. - Multi-dimension research (price + news combined) — use
gate-info-research.
Workflow
Step 1 — Major events (last 24–72h)
# Use web_search (general tool) for recent crypto news
web_search("crypto major news last 24 hours", recency="day", limit=10)
Step 2 — Trending stories
web_search("trending crypto stories today", recency="day", limit=10)
Step 3 — Social sentiment
Search X/Twitter and major crypto forums for the top 3 stories; classify sentiment (bullish/bearish/neutral) per story from headline + reply tone.
Step 4 — Synthesize
## News Briefing (playbook: news_briefing)
### Major Events
- [event] — [1-line impact]
### Trending
- [story] — [why it's trending]
### Sentiment
- Overall: bullish / neutral / bearish (with the 1–2 drivers)
### Sources
- web_search queries + named outlets
Anti-Rationalization Table
| Rationalization | Reality |
|---|---|
| "One headline = the story" | Triangulate 2–3 sources before stating impact. |
| "Trending = important" | Trending is attention; weight by source credibility. |
| "Sentiment is a price call" | Sentiment is a read on attention/mood, not a forecast. |
Code Example
Pull recent crypto news with the general web tool:
web_search("crypto market news today", recency="day", limit=8)
# → [ {title:"ETF inflows resume", url:...}, {title:"L2 TVL dips", url:...}, ... ]
Verification
- Major events pulled for the stated window (24–72h) via web_search.
- Trending stories separated from major events.
- Social sentiment classified per top story with drivers named.
- Output follows the News Briefing template (Major / Trending / Sentiment / Sources).
- Sources named (queries + outlets); no unattributed claims.