install
source · Clone the upstream repo
git clone https://github.com/henry-y/openclaw-paper-tools
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/henry-y/openclaw-paper-tools "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/hf-daily-papers" ~/.claude/skills/henry-y-openclaw-paper-tools-hf-daily-papers && rm -rf "$T"
OpenClaw · Install into ~/.openclaw/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/henry-y/openclaw-paper-tools "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/skills/hf-daily-papers" ~/.openclaw/skills/henry-y-openclaw-paper-tools-hf-daily-papers && rm -rf "$T"
manifest:
skills/hf-daily-papers/SKILL.mdsource content
HF Daily Papers (OpenClaw Skill)
Generate a daily list of trending papers from Hugging Face Papers, filtered into:
- Generation (diffusion / video / image / generative)
- Efficient (attention / sparse / quant / memory / optimization)
Outputs:
- Markdown:
skills/hf-daily-papers/recommendations/YYYY-MM-DD.md - (Optional) PDF:
skills/hf-daily-papers/recommendations/YYYY-MM-DD.pdf
Quick Use (chat)
- "Generate today's HF papers"
- "HF daily papers"
- "Generate HF daily papers and send me the list"
CLI
cd skills/hf-daily-papers python3 generator.py # PDF (requires `fpdf`) python3 generator.py --pdf
Environment Variables
(optional): HTTP(S) proxy URL, e.g.HF_DAILY_PAPERS_PROXYhttp://127.0.0.1:7897
Notes
- This skill scrapes
to collect up to 30 recent paper IDs, then fetches details viahttps://huggingface.co/papers
.https://huggingface.co/api/papers/<id>