report-writing-skills

Automates collecting, downloading, and analyzing recent research reports with customizable focus and real-time progress feedback.

This skill appears to implement web login + scraping and local report generation, which matches its description — but there are some red flags you should consider before installing or running it: 1) Credentials handling: The skill needs website credentials to log in, yet the registry declares no required environment variables or primary credential. Ask the author how credentials are supplied and stored. Prefer supplying credentials interactively or via a scoped secret (env var) rather than letting code write them to disk. 2) Dependencies: SKILL.md mentions Playwright and a headless browser but there is no install spec. Ensure the runtime environment has the correct Python packages (Playwright, tqdm) and that Playwright is installed and configured safely. Installing Playwright often downloads browser binaries — understand where those will be stored. 3) Filesystem paths: Some scripts use absolute paths under /root/.openclaw/... and create files on Desktop by default. Review and, if needed, change default output directories so nothing is written to privileged or unexpected locations. 4) Network & terms: This skill automates logging into and scraping third-party websites. Confirm you have permission to scrape the target site (terms of service, robots.txt), and avoid supplying credentials for sites you don't control. 5) Interactive vs non-interactive: report_writing_oil.py contains a non-interactive demo that hard-codes output locations and simulates downloads. If you plan to run this skill non-interactively (autonomously), inspect and test the scripts in a safe environment (non-production account) to see actual network behavior. 6) Code audit: Because the package contains runnable Python scripts, review the code (especially any network calls, subprocess invocations, or unexpected sys.path manipulations) before granting it access to real credentials or running it with elevated privileges. If you want to proceed safely: run the skill in an isolated environment (container or VM), provide credentials only via a temporary, least-privilege account, and verify network traffic and filesystem writes during a test run. If the author can update the skill to declare required credentials and dependencies explicitly and remove hard-coded root paths, that would make the package more coherent and easier to trust.

Type: OpenClaw Skill Name: report-writing-skills Version: 1.0.3 The skill bundle provides a legitimate framework for automating the collection and analysis of research reports from the website sgpjbg.com. The Python scripts (report_writing_workflow_v103.py and report_writing_oil.py) implement standard file handling, interactive user input for save locations, and progress tracking using tqdm. There is no evidence of data exfiltration, malicious execution, or prompt injection; the code behavior is consistent with the stated purpose of generating market analysis reports.