gdelt
starsforks
How to install gdelt?
kansoku-trade/kansoku/gdelt
$npx -y skills add kansoku-trade/kansoku --skill gdeltGlobal multilingual news event stream with tone scoring via GDELT 2.0 Doc API.
SKILL.md
| 1 | # gdelt |
| 2 | |
| 3 | > Response language: match user input. |
| 4 | |
| 5 | > ⚠️ **GDELT is a rolling recent-window API**, not a historical event archive. |
| 6 | > Time windows are anchored to absolute timestamps (UTC) for reproducibility — |
| 7 | > the same query asked tomorrow will return different results. |
| 8 | > |
| 9 | > ⚠️ **5-second throttle** between requests (enforced). Plan batches accordingly. |
| 10 | |
| 11 | ## When to use |
| 12 | |
| 13 | Trigger phrases: |
| 14 | |
| 15 | - 全球新闻 / 全球事件 / 多语种新闻 |
| 16 | - 媒体 tone / sentiment trend / 国际关系 |
| 17 | - geopolitical / event tone |
| 18 | - GDELT |
| 19 | |
| 20 | Useful for "what is the world saying about X right now" — i.e. retrieving |
| 21 | articles from non-English / non-financial sources that don't surface in |
| 22 | Longbridge's curated newsfeed. |
| 23 | |
| 24 | ## Workflow |
| 25 | |
| 26 | 1. Build a query in GDELT DSL (the user's term, optionally with operators like |
| 27 | `domain:bloomberg.com`, `sourcelang:eng`). |
| 28 | 2. Pick a mode: |
| 29 | - `artlist` — list of articles (default). |
| 30 | - `timelinetone` — per-15-min tone time series (-10 = very negative, |
| 31 | +10 = very positive). |
| 32 | - `timelinevol` / `timelinevolinfo` — article volume over time. |
| 33 | - `tonechart` — tone histogram. |
| 34 | 3. Specify the window — prefer `--start`/`--end` (absolute), fall back to |
| 35 | `--timespan`. The script converts relative timespans to absolute timestamps |
| 36 | before the call and echoes them in `meta.window` so the journal can be |
| 37 | re-run. |
| 38 | |
| 39 | ## CLI examples |
| 40 | |
| 41 | ```bash |
| 42 | # Articles about Nvidia in the last 24h |
| 43 | python3 .claude/skills/gdelt/scripts/doc.py "Nvidia" |
| 44 | |
| 45 | # 7-day window, English + Chinese articles about TSMC |
| 46 | python3 .claude/skills/gdelt/scripts/doc.py "TSMC OR \"Taiwan Semiconductor\"" --timespan 7d --lang eng,zho |
| 47 | |
| 48 | # Tone timeline for Federal Reserve over 30 days |
| 49 | python3 .claude/skills/gdelt/scripts/doc.py "Federal Reserve" --mode timelinetone --timespan 30d |
| 50 | |
| 51 | # Absolute window |
| 52 | python3 .claude/skills/gdelt/scripts/doc.py "AI chips" --start 20260501000000 --end 20260528000000 |
| 53 | ``` |
| 54 | |
| 55 | ## Output shape (artlist) |
| 56 | |
| 57 | ```json |
| 58 | { |
| 59 | "data": [ |
| 60 | { |
| 61 | "url": "https://...", |
| 62 | "title": "...", |
| 63 | "seendate": "20260527T161500Z", |
| 64 | "domain": "...", |
| 65 | "language": "English", |
| 66 | "sourcecountry": "United States", |
| 67 | "socialimage": "..." |
| 68 | } |
| 69 | ], |
| 70 | "meta": { |
| 71 | "mode": "artlist", |
| 72 | "query": "Nvidia", |
| 73 | "window": { "start": "20260527071804", "end": "20260528071804" }, |
| 74 | "max_records": 75 |
| 75 | }, |
| 76 | "ok": true |
| 77 | } |
| 78 | ``` |
| 79 | |
| 80 | ## Output shape (timelinetone) |
| 81 | |
| 82 | ```json |
| 83 | { |
| 84 | "ok": true, |
| 85 | "data": [ |
| 86 | {"date": "20260520T000000Z", "value": 1.42}, |
| 87 | {"date": "20260520T001500Z", "value": 1.05}, |
| 88 | ... |
| 89 | ], |
| 90 | "meta": {"mode": "timelinetone", ...} |
| 91 | } |
| 92 | ``` |
| 93 | |
| 94 | ## Error handling |
| 95 | |
| 96 | | Exit code | Meaning | LLM action | |
| 97 | | --------- | --------------------------------------- | ------------------------------------------------------------------------------------ | |
| 98 | | 0 | Success | Parse `data`. | |
| 99 | | 1 | Invalid args (e.g. bad timespan / lang) | Read `hint`. | |
| 100 | | 3 | HTTP 4xx / non-JSON response | If body contains "Please limit requests", the throttle was tripped — wait and retry. | |
| 101 | | 4 | Network | Suggest retry. | |
| 102 | |
| 103 | ## Known limitations |
| 104 | |
| 105 | - 5-second minimum between requests; batch tone + artlist queries must be |
| 106 | sequenced. |
| 107 | - `--max-records` cap is 250. |
| 108 | - GDELT's tone metric is a heuristic — useful for direction-of-narrative, not |
| 109 | ground truth. |
| 110 | - Results are not cached (window-sensitive). |
| 111 | |
| 112 | ## Related skills |
| 113 | |
| 114 | - `longbridge-news` for curated equity-specific newsfeed (Chinese-language UX). |
| 115 | - `sec-edgar` for primary-source filings as the contrast to media narrative. |
| 116 | - `fred` for macro data referenced in the narrative. |
People who liked this also viewed
6 picksType
- azure-costAzure cost management: query costs, forecast spending, optimize to reduce waste. WHEN: \"Azure costs\", \"Azure bill\", \"cost breakdown\", \"how much am I spending\", \"forecast spending\", \"optimizSkillsJul 2026
- pricingWhen the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increaseSkillsJul 2026
- stripe-integrationMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- startup-financial-modelingMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- tushare-finance专为 Claude Code 设计的 Tushare Pro 金融数据技能包，支持 220+ 个接口，用自然语言获取 A 股行情、财务数据、宏观经济等金融数据。SkillsJul 2026
- azure-pricingFetches real-time Azure retail pricing using the Azure Retail Prices API (prices.azure.com) and estimates Copilot Studio agent credit consumption. Use when the user asks about the cost of any Azure seSkillsJul 2026
