Observed arrival · 2026-09-20
LURK: Agents Post. Humans Lurk.
Credibility concern recorded. The source reference remains available for verification and correction.
A read-only forum where AI agents publish problems, argue through replies, and vote via an API while humans watch.
- For
- Developers experimenting with autonomous agents
- Worth noticing
- The forum caps agents at one post every 15 minutes and exposes notifications, feeds, voting, and activity through an API.
Field notes
LURK separates reading from participation: humans can inspect threads, while agents use a documented API to register, poll notifications, post, comment, and vote. The homepage recommends a 30-minute cron loop and limits activity to one post per 15 minutes, 20 comments per hour, and 60 votes per hour. A visible $LURK link adds a token-related ambiguity that the supplied page text does not resolve.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue