From d8a6bf08970beba2c5431d61c401e11be645526a Mon Sep 17 00:00:00 2001 From: willamhou Date: Thu, 5 Mar 2026 13:26:25 +0800 Subject: [PATCH] docs: arxiv-reader now works standalone (no Docker required) --- usecases/arxiv-paper-reader.md | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/usecases/arxiv-paper-reader.md b/usecases/arxiv-paper-reader.md index 35c7b9f..f606c1c 100644 --- a/usecases/arxiv-paper-reader.md +++ b/usecases/arxiv-paper-reader.md @@ -13,17 +13,14 @@ This workflow turns your agent into a research reading assistant: ## Skills you Need - [arxiv-reader](https://github.com/Prismer-AI/Prismer/tree/main/skills/arxiv-reader) skill (3 tools: `arxiv_fetch`, `arxiv_sections`, `arxiv_abstract`) -- Prismer workspace container (runs the arXiv server on port 8082) + +No Docker or Python required — the skill runs standalone using Node.js built-ins. It downloads directly from arXiv, decompresses the LaTeX source, and flattens includes automatically. ## How to Set it Up -1. Clone and deploy [Prismer](https://github.com/Prismer-AI/Prismer) with Docker (the arXiv server starts automatically): -```bash -git clone https://github.com/Prismer-AI/Prismer.git && cd Prismer -docker compose -f docker/docker-compose.dev.yml up -``` +1. Install the `arxiv-reader` skill from the [Prismer repository](https://github.com/Prismer-AI/Prismer/tree/main/skills/arxiv-reader) — copy the `skills/arxiv-reader/` directory into your OpenClaw skills folder. -2. The `arxiv-reader` skill is built-in — no installation needed. Prompt OpenClaw: +2. The skill is ready to use. Prompt OpenClaw: ```text I'm researching [topic]. Here's my workflow: