Lay Press Release Writer

Security checks across static analysis, malware telemetry, and agentic risk

Overview

The skill is mostly local and purpose-aligned, but it appears to invent attributed researcher quotes and media contact details for public press releases.

Use this skill only as a drafting aid. Verify the paper summary, replace any generated quotes and media contacts with approved real information, and review the output carefully before publishing.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

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 publish a press release containing quotes or contact information that appear real but were actually invented, creating reputational, ethical, or factual risk.

Why it was flagged

The script constructs attributed researcher quotes and a media-contact email from names/institution text instead of extracting or requiring verified source material.

Skill content
quotes.append(f"\"This research provides a new perspective on our understanding of the field,\"{first_author}English,\"We expect this discovery to bring practical value to related applications。\"") ... "email": f"media@{institution.lower().replace('University', '').replace('college', '')}.edu.cn"
Recommendation

Treat all generated quotes and contact details as placeholders; require user-supplied, verified, and approved quotes/contact information before publication.

What this means

The agent may execute the packaged Python script on the local machine to generate the press release.

Why it was flagged

The skill is intended to run a local Python script. This is disclosed and purpose-aligned, but it is still local code execution.

Skill content
python scripts/main.py --paper-text "Paper content..." --institution "XX University"
Recommendation

Run it only in a normal workspace with reviewed inputs and inspect the generated output before using it externally.