report-writing-skills

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill is meant to collect research reports, but some included scripts appear to simulate authenticated downloads and generate hard-coded results, so users should review it before relying on its reports.

Use this skill cautiously: verify that it really logs in, searches, and downloads the intended reports before trusting the generated analysis, and provide only credentials for the specific research site you intend to use.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

A user could believe the final analysis is based on freshly authenticated downloads when it may be based on hard-coded or placeholder information.

Why it was flagged

The script reports successful authentication and workflow progress while explicitly using a simulated, non-interactive path and saving metadata-only placeholders rather than downloaded PDFs.

Skill content
# Simulate workflow with hard-coded requirements (non-interactive) ... print("✅ 网站认证登录成功") ... f.write(f"Status: Metadata only (PDF download requires authentication)\n")
Recommendation

Do not rely on generated reports without verifying the source documents were actually downloaded and reviewed; the maintainer should clearly label demo/simulation code or replace it with real, user-confirmed collection logic.

What this means

The agent may ask for or use your research-site account credentials during the workflow.

Why it was flagged

The skill is expected to use user-provided login credentials for a research website, which is aligned with its stated report-download purpose but still grants delegated account access.

Skill content
1. **网站认证** - 使用提供的凭据登录研究网站
Recommendation

Use only credentials for the intended research site, avoid sharing broader account access, and prefer a limited or disposable account if available.

What this means

The agent may browse a site, trigger downloads, and create report files on your machine.

Why it was flagged

The workflow automates external website interaction, downloads, and local file writes; these are expected for report collection but are still actions the user should confirm.

Skill content
点击下载按钮或访问下载链接
- 将文件保存到指定目录
Recommendation

Confirm the target website, selected reports, and save directory before allowing the workflow to run.

What this means

If run in a different environment, the script could fail or import code from an unexpected local path.

Why it was flagged

A helper script prepends a hard-coded local path for imports instead of using its own package-relative location, creating ambiguity about exactly which local module would be loaded if that path differs from the reviewed files.

Skill content
sys.path.insert(0, '/root/.openclaw/workspace/skills/report-writing-v1.03/scripts')
Recommendation

Use package-relative imports and verify the installed script directory before executing helper scripts.