dpickem

Systematically investigate a bug or unexpected behaviour to identify the root cause

Improve the structure of existing code without changing its external behaviour

Use a notes file in your workspace as working memory during a task

Perform a structured, constructive code review of the provided changes

Summarize arXiv paper intelligence digests for ML, AI, foundation/foundation models, LLMs, post-training, agentic workflows, evals, harness engineering, loop engineering, recursive self-improvement, and self-improving systems. Use when asked to run or interpret the loopcraft arXiv workflow, prepare a daily paper digest, or identify the most interesting recent arXiv submissions.

Triage and summarize everything aimed at the operator on Slack — @mentions, DMs, group DMs, and a configured list of high-signal channels — into a single digest grouped by theme, led by a high-level topic overview. Observe-only: never send a message.

Summarize X/Twitter intelligence digests for ML, AI, frontier/foundation model practice, agents, evals, tool use, harness engineering, loop engineering, and loopcraft. Use when asked to run or interpret the loopcraft X intelligence workflow, prepare a daily Codex automation report from X posts, highlight frontier-lab posts, discover follow candidates, or extract direct links from X posts and linked external articles/blogs/papers.