Under the hood

AI queue

Watch AgentDish crawl new pages and run AI reviews in public.

Processing queue

Live queue state for crawls and AI evaluations.

#3 submission_evaluation succeeded

https://github.com/evanklem/evanflow

complete 2026-04-27 23:53:54

#2 submission_evaluation succeeded

https://github.com/dirac-run/dirac

complete 2026-04-27 23:53:48

#1 submission_evaluation succeeded

https://agentswarms.fyi

complete 2026-04-27 07:09:15

Processing logs

Newest worker events first.

info / complete Processing job completed. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:31

info / screenshot_captured Screenshot captured and linked. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:31

info / capturing_screenshot Capturing screenshot for QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:28

info / complete Accepted and published QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:28

info / publishing Publishing accepted listing. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:28

info / evaluation_saved Saved accepted evaluation with score 83. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:28

info / evaluating Running the AI evaluation. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:21

info / crawl_saved Saved crawl snapshot with status 200. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:21

info / crawling Crawling https://arxiv.org/abs/2605.24218. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:21

info / checking_duplicate Checking for an existing accepted listing. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:21

info / starting Started processing job. https://arxiv.org/abs/2605.24218 / 2026-05-26 15:33:21

info / complete Processing job completed. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:19

info / screenshot_captured Screenshot captured and linked. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:19

info / capturing_screenshot Capturing screenshot for Building the harness around our coding agents: eight failure modes, eight pillars. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:15

info / complete Accepted and published Building the harness around our coding agents: eight failure modes, eight pillars. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:15

info / publishing Publishing accepted listing. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:15

info / evaluation_saved Saved accepted evaluation with score 74. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:15

info / evaluating Running the AI evaluation. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:12

info / crawl_saved Saved crawl snapshot with status 200. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:12

info / crawling Crawling https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/. https://nimbalyst.com/blog/agent-harness-above-claude-code-codex/ / 2026-05-26 15:33:11