sodalone/paper_collector

Latest recorded source activity: September 21, 2026 at 02:43

SkillsMP catalog refreshed: November 9, 2026 at 14:22

skills collected: 1

GitHub stars: 0
GitHub forks: 0

Install with an AI assistant

Copy this prompt into the AI assistant you're using.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/sodalone/paper_collector.

Repository README

Repository README source: README.md

Read the source README

Expand full README

# Paper Collector 面向 AI 论文监控的本地 arXiv 收集仓库。当前核心能力是通过 `collect-ai-arxiv` Skill 从 arXiv `cs.RO` 和 `cs.CV` 抓取论文，按不同 profile 分类，并生成中文 Markdown 日报或周报。 ## 能力范围 - `embodied`: 具身智能论文日报/周报。 - `autonomous-driving`: 自动驾驶论文日报/周报。 - 默认 source 固定为 `cs.RO,cs.CV`。 - 默认按 Asia/Shanghai 自然日或自然周生成报告。 - 输出 Markdown 报告，并保留本地 JSON、raw、分类缓存等可再生成中间产物。 ## 目录结构 ```text collect-ai-arxiv/ ├── SKILL.md ├── agents/openai.yaml ├── references/ │ ├── report-template.md │ └── taxonomy.md └── scripts/collect_arxiv.py reports/ ├── embodied/ └── autonomous-driving/ tests/ └── test_collect_arxiv.py ``` `reports/**/*.md` 是可提交的报告正文；`papers.json`、`raw/` 和 `reports/.cache/` 是可再生成的本地中间产物，已通过 `.gitignore` 忽略。 ## 使用方式从仓库根目录运行： ```bash python3 collect-ai-arxiv/scripts/collect_arxiv.py daily --profile embodied --out ./reports python3 collect-ai-arxiv/scripts/collect_arxiv.py weekly --profile embodied --out ./reports python3 collect-ai-arxiv/scripts/collect_arxiv.py daily --profile autonomous-driving --out ./reports python3 collect-ai-arxiv/scripts/collect_arxiv.py weekly --profile autonomous-driving --out ./reports ``` 指定日期或周： ```bash python3 collect-ai-arxiv/scripts/collect_arxiv.py daily --profile embodied --date 2026-06-24 --out ./reports python3 collect-ai-arxiv/scripts/collect_arxiv.py weekly --profile autonomous-driving --week 2026-W25 --out ./reports ``` 默认分类器会调用本机 `codex exec`。需要快速离线草稿时可以使用规则分类器： ```bash python3 collect-ai-arxiv/scripts/collect_arxiv.py weekly --profile autonomous-driving --week 2026-W25 --out ./reports --classifier rules ``` ## 报告命名 Markdown 文件名与报告标题保持一致： ```text reports/embodied/daily/具身智能 arXiv YYYY-MM-DD 日报.md reports/embodied/weekly/具身智能 arXiv YYYY-Www-MMDD-MMDD 周报.md reports/autonomous-driving/daily/自动驾驶 arXiv YYYY-MM-DD 日报.md reports/autonomous-driving/weekly/自动驾驶 arXiv YYYY-Www-MMDD-MMDD 周报.md ``` 中间产物继续使用短 label bundle： ```text reports/embodied/daily/YYYY-MM-DD/papers.json reports/autonomous-driving/weekly/YYYY-Www/raw/ ``` ## 验证 ```bash python3 -m unittest tests/test_collect_arxiv.py python3 -m py_compile collect-ai-arxiv/scripts/collect_arxiv.py python3 ~/.codex/skills/.system/skill-creator/scripts/quick_validate.py collect-ai-arxiv ``` ## 设计原则 - 分类先看论文核心贡献，不按关键词顺序抢主类。 - 具身报告正文以 `Embodied Task` 索引为唯一正文入口。 - 自动驾驶报告正文以 `Driving Stack` 索引为唯一正文入口。 - 可再生成的抓取结果、raw cache 和分类 cache 不进版本库。

Loading the repository README…

Skills in this repository

classification pending

skill

occupation

description

updated

collect-ai-arxiv

unclassified

Collect and classify AI papers from arXiv cs.RO and cs.CV with profile-specific Chinese daily or weekly Markdown reports. Use when the user asks for 具身智能 arXiv 论文收集、自动驾驶 arXiv 日报/周报、论文分类、VLA/robotics/embodied AI monitoring, autonomous-driving paper…

Source text: Chinese

2026-09-21

Showing 1 of 1 collected skill.