mirror of
https://github.com/koala73/worldmonitor.git
synced 2026-04-25 17:14:57 +02:00
5cec1b8c4c
* fix(insights): trust cluster rank, stop LLM from re-picking top story WORLD BRIEF panel published "Iran's new supreme leader was seriously wounded, leading him to delegate power to the Revolutionary Guards. This development comes amid an ongoing war with Israel." to every visitor for 3h. Payload: openrouter / gemini-2.5-flash. Root cause: callLLM sent all 10 clustered headlines with "pick the ONE most significant and summarize ONLY that story". Clustering ranked Lebanon journalist killing #1 (2 corroborating sources); News24 Iran rumor ranked #3 (1 source). Gemini overrode the rank, picked #3, and embellished with war framing from story #4. Objective rank (sourceCount, velocity, isAlert) lost to model vibe. Shrink the LLM's job to phrasing. Clustering already ranks — pass only topStories[0].primaryTitle and instruct the model to rewrite it using ONLY facts from the headline. No name/place/context invention. Also: - temperature 0.3 -> 0.1 (factual summary, not creative) - CACHE_TTL 3h -> 30m so a bad brief ages out in one cron cycle - Drop dead MAX_HEADLINES const Payload shape unchanged; frontend untouched. * fix(insights): corroboration gate + revert TTL + drop unconditional WHERE Follow-up to review feedback on the ranking contract, TTL, and prompt: 1. Corroboration gate (P1a). scoreImportance() in scripts/_clustering.mjs is keyword-heavy (violence +125 on a single word, flashpoint +75, ^1.5 multiplier when both hit), so a single-source sensational rumor can outrank a 2-source lead purely on lexical signals. Blindly trusting topStories[0] would let the ranker's keyword bias still pick bad stories. Walk topStories for sourceCount >= 2 instead — corroboration becomes a hard requirement, not a tiebreaker. If no cluster qualifies, publish status=degraded with no brief (frontend already handles this). 2. CACHE_TTL back to 10800 (P1b). 30m TTL == one cron cadence means the key expires on any missed or delayed run and /api/bootstrap loses insights entirely (api/bootstrap.js reads news:insights:v1 directly, no LKG across TTL-gap). The short TTL was defense-in-depth for bad content; the real safety is now upstream (corroboration gate + grounded prompt), so the LKG window doesn't need to be sacrificed for it. 3. Prompt: location conditional (P2). "Use ONLY facts present" + "Lead with WHAT happened and WHERE" conflicted for headlines without an explicit location and pushed the model toward inferred-place hallucination. Replaced with "Include a location, person, or organization ONLY if it appears in the headline." * test(insights): lock corroboration gate + grounded-prompt invariants Review P2: the corroboration gate and the prompt's no-invention rules had no tests, so future edits to selectTopStories() ordering or prompt text could silently reintroduce the original hallucination. Extract the brief-selection helper and prompt builders into a pure module (scripts/_insights-brief.mjs) so tests can import them without triggering seed-insights.mjs's top-level runSeed() call: - pickBriefCluster(topStories) returns first sourceCount>=2 cluster - briefSystemPrompt(dateISO) returns the system prompt - briefUserPrompt(headline) returns the user prompt Regression tests (tests/seed-insights-brief.test.mjs, 12 cases) lock: - pickBriefCluster skips single-source rumors even when ranked above a multi-sourced lead (explicit regression: News24 Iran supreme leader 2026-04-23 scenario with realistic scores) - pickBriefCluster tolerates missing/null entries - briefSystemPrompt forbids invented facts and proper nouns - briefSystemPrompt's "location" rule is conditional (no unconditional "Lead with WHAT and WHERE" directive that would push the model toward place-inference when the headline has no location) - briefSystemPrompt does not contain "pick the most important" style language (ranking is done by pickBriefCluster upstream) - briefUserPrompt passes the headline verbatim and instructs "only facts from this headline" Also fix a misleading comment on CACHE_TTL: corroboration is gated at brief-selection time, not on the topStories payload itself (which still includes single-source clusters rendered as the headline list). test:data: 6657/6657 pass (was 6645; +12).
World Monitor
Real-time global intelligence dashboard — AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface.
-0078D4?style=for-the-badge&logo=windows&logoColor=white){kind=icon}
-FCC624?style=for-the-badge&logo=linux&logoColor=black){kind=icon}
Documentation · Releases · Contributing
What It Does
- 500+ curated news feeds across 15 categories, AI-synthesized into briefs
- Dual map engine — 3D globe (globe.gl) and WebGL flat map (deck.gl) with 45 data layers
- Cross-stream correlation — military, economic, disaster, and escalation signal convergence
- Country Intelligence Index — composite risk scoring across 12 signal categories
- Finance radar — 92 stock exchanges, commodities, crypto, and 7-signal market composite
- Local AI — run everything with Ollama, no API keys required
- 5 site variants from a single codebase (world, tech, finance, commodity, happy)
- Native desktop app (Tauri 2) for macOS, Windows, and Linux
- 21 languages with native-language feeds and RTL support
For the full feature list, architecture, data sources, and algorithms, see the documentation.
Quick Start
git clone https://github.com/koala73/worldmonitor.git
cd worldmonitor
npm install
npm run dev
Open localhost:5173. No environment variables required for basic operation.
For variant-specific development:
npm run dev:tech # tech.worldmonitor.app
npm run dev:finance # finance.worldmonitor.app
npm run dev:commodity # commodity.worldmonitor.app
npm run dev:happy # happy.worldmonitor.app
See the self-hosting guide for deployment options (Vercel, Docker, static).
Tech Stack
| Category | Technologies |
|---|---|
| Frontend | Vanilla TypeScript, Vite, globe.gl + Three.js, deck.gl + MapLibre GL |
| Desktop | Tauri 2 (Rust) with Node.js sidecar |
| AI/ML | Ollama / Groq / OpenRouter, Transformers.js (browser-side) |
| API Contracts | Protocol Buffers (92 protos, 22 services), sebuf HTTP annotations |
| Deployment | Vercel Edge Functions (60+), Railway relay, Tauri, PWA |
| Caching | Redis (Upstash), 3-tier cache, CDN, service worker |
Full stack details in the architecture docs.
Flight Data
Flight data provided gracefully by Wingbits, the most advanced ADS-B flight data solution.
Data Sources
WorldMonitor aggregates 65+ external data sources across geopolitics, finance, energy, climate, aviation, cyber, military, infrastructure, and news intelligence. See the full data sources catalog for providers, feed tiers, and collection methods.
Contributing
Contributions welcome! See CONTRIBUTING.md for guidelines.
npm run typecheck # Type checking
npm run build:full # Production build
License
AGPL-3.0 for non-commercial use. Commercial license required for any commercial use.
| Use Case | Allowed? |
|---|---|
| Personal / research / educational | Yes |
| Self-hosted (non-commercial) | Yes, with attribution |
| Fork and modify (non-commercial) | Yes, share source under AGPL-3.0 |
| Commercial use / SaaS / rebranding | Requires commercial license |
See LICENSE for full terms. For commercial licensing, contact the maintainer.
Copyright (C) 2024-2026 Elie Habib. All rights reserved.
Author
Elie Habib — GitHub
Contributors
Security Acknowledgments
We thank the following researchers for responsibly disclosing security issues:
- Cody Richard — Disclosed three security findings covering IPC command exposure, renderer-to-sidecar trust boundary analysis, and fetch patch credential injection architecture (2026)
See our Security Policy for responsible disclosure guidelines.
worldmonitor.app · docs.worldmonitor.app · finance.worldmonitor.app · commodity.worldmonitor.app