agent-survey-corpus
installsforks
How to install agent-survey-corpus?
willoscar/research-units-pipeline-skills/agent-survey-corpus
$npx -y skills add WILLOSCAR/research-units-pipeline-skills --skill agent-survey-corpusDownload a small corpus of open-access arXiv survey/review PDFs about agentic systems and extract text for style learning. Trigger: agent survey corpus, ref corpus, download surveys, 学习综述写法, 下载 survey. Use when: you want to study how real agent surveys structure sections
SKILL.md
| 1 | # Agent Survey Corpus (arXiv PDFs → text extracts) |
| 2 | |
| 3 | Goal: create a small, local reference library so you can **learn from real agent surveys** when refining: |
| 4 | - C2 outline structure (paper-like sectioning) |
| 5 | - C4 tables/claims organization |
| 6 | - C5 writing style and density |
| 7 | |
| 8 | This is intentionally *not* part of the pipeline; it is an optional, repo-level toolkit. |
| 9 | |
| 10 | ## Inputs |
| 11 | |
| 12 | - `ref/agent-surveys/arxiv_ids.txt` |
| 13 | |
| 14 | ## Outputs |
| 15 | |
| 16 | - `ref/agent-surveys/pdfs/` |
| 17 | - `ref/agent-surveys/text/` |
| 18 | - `ref/agent-surveys/STYLE_REPORT.md` (tracked; auto-generated summary) |
| 19 | |
| 20 | ## Workflow |
| 21 | |
| 22 | 1) Edit `ref/agent-surveys/arxiv_ids.txt` (one arXiv id per line). |
| 23 | 2) Run the downloader to fetch PDFs and extract the first N pages to text. |
| 24 | 3) Skim the extracted text under `ref/agent-surveys/text/`: |
| 25 | - look at section counts (H2), subsection granularity (H3), and how they transition between chapters. |
| 26 | - identify repeated rhetorical patterns you want the pipeline writer to imitate. |
| 27 | |
| 28 | ## Script |
| 29 | |
| 30 | ### Quick Start |
| 31 | |
| 32 | - `uv run python .codex/skills/agent-survey-corpus/scripts/run.py --help` |
| 33 | - `uv run python .codex/skills/agent-survey-corpus/scripts/run.py --workspace . --max-pages 20` |
| 34 | |
| 35 | ### All Options |
| 36 | |
| 37 | - `--workspace <dir>` (use `.` to write into repo root) |
| 38 | - `--inputs <semicolon-separated>` (default: `ref/agent-surveys/arxiv_ids.txt`) |
| 39 | - `--max-pages <N>` (default: 20) |
| 40 | - `--sleep <seconds>` (default: 1.0) |
| 41 | - `--overwrite` (re-download + re-extract) |
| 42 | |
| 43 | ### Examples |
| 44 | |
| 45 | - Download/extract into repo root `ref/`: |
| 46 | - `uv run python .codex/skills/agent-survey-corpus/scripts/run.py --workspace . --max-pages 20` |
| 47 | - Download/extract into a specific folder (treated as workspace root): |
| 48 | - `uv run python .codex/skills/agent-survey-corpus/scripts/run.py --workspace /tmp/surveys --max-pages 30` |
| 49 | |
| 50 | ## Troubleshooting |
| 51 | |
| 52 | - **Download fails / timeout**: rerun with a larger `--sleep`, or try fewer ids. |
| 53 | - **Text extract is empty**: the PDF may be scanned; try another survey or increase `--max-pages`. |
| 54 | - **Files showing up in git status**: PDFs/text are ignored via `.gitignore` (`ref/**/pdfs/`, `ref/**/text/`). |
People who liked this also viewed
6 picksType
- find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 2026
- grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026
- vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026
- improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026
- setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026
- triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026
