Admin

Admin

Manage sources, companies, and monitor ingestion.

What do these buttons do?
Run Ingestion — Crawl all active sources and add new articles to the database.
Extract Entities — Scan new articles for company/person/topic mentions (regex-based, fast).
Gemini Enrich — Use Gemini AI to extract richer entities from high-scoring articles. Requires GEMINI_API_KEY.
Run Discovery — Find new sources, companies, and topics not yet tracked.
Taxonomy Discovery — Detect new vendors, associations, and roles; adds to Discovery page.
Sync Known Universe — Push accepted taxonomy candidates into the scoring engine.
Weekly Report — Summarise new and promoted entities for the week.
Relevance Cleanup — Delete articles below score 15 and re-score borderline ones.

Pipeline KPIs

Auto-ingestion: 0 */2 * * *
10
Articles today
93 last 7 days
99%
Crawl success
7412 runs last 7 days
72
Avg relevance score
across all indexed articles
90%
Clustered
articles assigned to a cluster
Top sources by volume
Google News: "chief communications officer"348
Google News: AMEC290
Google News: "Kommunikationschef"284
Google News: Cision241
Google News: "corporate communications"151

Infrastructure

59 MB
Database size
+26 KB/day est. at current rate
780 KB
Projected monthly growth
~26 KB/day × 30 · ~2 KB per article
3.5 GB
Projected monthly traffic
827.7 MB measured last 7d
858
Requests / day
80 sources · avg every 134 min
0%
Store rate (7d)
93 stored / 296,447 fetched — 100% filtered by score gate

Recent Crawl Runs

View all →
SourceStatusFoundNewStarted
Silicon Anglesuccess30010/2/2026, 8:00:38 AM
Prowly Magazinesuccess10010/2/2026, 8:00:38 AM
Google Alertsuccess0010/2/2026, 8:00:38 AM
Mention Blogsuccess9010/2/2026, 8:00:38 AM
Amec.orgsuccess12010/2/2026, 8:00:38 AM

Admin Sections