koala73 / koala73/worldmonitor
feat(mcp): tech-monitor feed exposure — Hacker News, GitHub trending, AI research papers via MCP
- Dominant language
- TypeScript
- Stars
- 86.6k
- Forks
- 13.1k
- Avg merge
- 8h 4m
- Merged PRs (30d)
- 825
Description
## Summary
The TECH monitor (tech.worldmonitor.app) aggregates technology-ecosystem signals, but the MCP surface only exposes a thin `get_research_signals` ("tech events bootstrap"). Agents asking "what's trending in AI research / on Hacker News / on GitHub this week" get little, even though this is exactly the audience our MCP serves.
## Proposed scope
1. Expose the tech-monitor feed corpus through MCP — either by widening `get_research_signals` with filters or adding a `get_tech_signals` tool:
- Hacker News top/trending stories
- GitHub trending repositories (language/time-window filters)
- AI research papers (arXiv cs.AI/cs.LG/cs.CL; add the feed if the tech variant doesn't already ingest it)
- Model-release tracking (HuggingFace trending) as a stretch item
2. Reuse the tech variant's existing seeders/caches wherever they exist; add missing sources as ordinary feed additions (`src/config/feeds.ts` conventions), not bespoke pipelines.
3. Standard cache-tool treatment: freshness, JMESPath, parity, docs.
## Existing assets
- Tech variant feeds + panels; `src/config/tech-companies.ts`, `ai-research-labs.ts`, `startup-ecosystems.ts` for entity context.
## Acceptance criteria
- [ ] HN + GitHub trending + research-paper signals queryable via MCP with filters.
- [ ] Sources shared with the tech variant (single ingestion path).
- [ ] Docs + parity entries shipped.
Contributor guide
Research direction
Start by tracing get_research_signals and the tech variant's feed setup in src/config/feeds.ts; inspect existing tech-variant seeders, caches, and the listed entity-context files. Done means HN, GitHub trending, and AI-paper signals are queryable with filters, share ingestion, and have freshness, JMESPath, parity, and documentation coverage.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- github, typescript
- Domain
- api, backend, data
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 38/100