Agent operations

Frontier lab agent command center

Operational state for all queue states: scrape runs, source health, the analysis queue, and model provenance.

last scrape 1h46/56 sources healthy18 analysesdeepseek · deepseek-v4-pro

Agent answer

Frontier lab agent operations show 18 queued organizations in all queue states: 17 runnable analyses, 0 page-capture targets, 0 scrape-repair targets, and 1 fresh monitors. DeepSeek v4 pro via deepagents is the dominant standing analysis path across 18 generated analysis rows. 10 of 56 sources need attention. Next dispatch: run_analysis_agent for Anthropic because new captured evidence is fresher than the stored analysis. Top signal inspection: NVIDIA - Skild AI Taps NVIDIA Physical AI to Teach Robots New Tasks From a Single Video with watch priority.

Run now17Needs pages0Fresh1No signals0source issues10
Run now17

Anthropic

npm run agent:analysis -- --url <app-url> --org <org-slug> --limit 1 --web-limit 10
Fresh1

Amazon (Nova)

Analysis queue

json →18 shown
All18Run now17Needs pages0Fresh1No signals0

Scrape runs

8 of 12snapshot
Sep 11, 2026
ok
duration 2m 8s · sources 32events 3,670 · HN 0failures clear
Sep 11, 2026
ok
duration 1m 54s · sources 32events 3,669 · HN 0failures clear
Sep 11, 2026
ok
duration 1m 42s · sources 32events 3,669 · HN 0failures clear
Sep 11, 2026
ok
duration 44s · sources 33events 823 · HN 0failures clear
Sep 11, 2026
failed
duration 4m 21s · sources 39events 1,492 · HN 0failures 4 · cerebras/careers, google-deepmind/careers, fireworks-ai/careers
Sep 11, 2026
ok
duration 1m 43s · sources 32events 3,669 · HN 0failures clear
Sep 11, 2026
ok
duration 2m 45s · sources 32events 3,669 · HN 1failures clear
Sep 10, 2026
failed
duration 2m 54s · sources 32events 3,568 · HN 4failures 1 · digitalocean/writing
4 older runs in snapshot

Source health

46 ok · 1 dead · 5 stale · 4 emptycoverage →