Content Summarizer

Fetch any URL and produce a structured content summary with extractive summarization, AI enhancement prompts, and structured output templates. Extracts clean...

This skill appears to do what it says: it fetches pages with curl, extracts readable text with simple Python routines, selects key sentences, and writes Markdown summaries to a local directory. Before installing, consider: 1) It will fetch arbitrary URLs you pass it — avoid summarizing sensitive internal URLs unless you understand the network exposure (it can be used to access internal services if run in a network with access). 2) It writes files under OUTPUT_DIR (default /tmp/summaries); change that if you want a different location and avoid running as root. 3) The SKILL.md mentions environment variables (MODEL, SUMMARIZER_MODEL, MAX_TOKENS) that aren’t declared in registry metadata — set these explicitly if you rely on them. 4) If you need stronger HTML parsing or handling of JS-heavy sites, this uses a simple regex-based extractor and may miss or garble content. 5) As with any script that fetches remote content, review and run it in a controlled environment if you have concerns about unintended network access.

Type: OpenClaw Skill Name: strd-content-summarizer Version: 1.0.0 The skill is a legitimate content summarization tool that fetches URLs via curl and processes text using Python regex for cleaning and extractive summarization. The scripts (url-to-summary.sh, summarize.sh, and key-points.sh) perform their stated functions without any evidence of data exfiltration, unauthorized execution, or malicious prompt injection. All outputs are saved locally to /tmp/summaries/ as intended.