QuantumBFS

quantum-bfs

sci-brain

Download academic papers as PDFs given a URL, DOI, title, or arXiv ID. Tries open web → Sci-Hub → arXiv.

Submit GPU compute jobs to a Slurm cluster: sbatch script generation, GPU type selection, log management.

Verify BibTeX references against CrossRef metadata. Finds uncited entries and flags metadata discrepancies.

Review a LaTeX research-paper draft against a fixed checklist and insert inline author-comment reminders.

Mine your local Claude Code session history for recurring behavioral feedback (corrections, format/voice edicts, error catches, domain-knowledge injections) and distill each pattern into a proposed CLAUDE.md addition, an auto-memory feedback entry, or — only for true workflow patterns — a SKILL.md draft.

Track motivation, watch criteria, and decision rules for long-running cluster jobs in a single BABYSIT.md file on the cluster — so monitoring state survives across sessions, machines, and conversation resets.

Extract numeric (x, y) data from a plot/figure image (or a PDF figure) into CSV — a CLI alternative to WebPlotDigitizer. Wraps plotdigitizer with PDF rendering, color isolation, and overlay verification.

Recover context from Claude Code session transcripts: locate a session by hash/UUID prefix or project path, mine ~/.claude/projects JSONL logs for handoff notes, summaries, and next steps, and print an approximate resume command. Ships a dependency-free streaming helper safe for multi-MB transcripts.

Download a video lecture from koushare.com (蔻享学术). The page sits behind a Tencent EdgeOne bot/CAPTCHA wall, but the video is a token-signed Tencent VOD HLS stream (voddrm.token, SimpleAES AES-128) — capture that CDN URL via a real browser and decrypt it with yt-dlp/ffmpeg.

Scientific research brainstorming partner: survey the literature, find good problems, and shape concrete research ideas together