Security audit
Security checks for vulnerabilities and agentic risk
ClawScan rates this skill benign with medium confidence: its reported purpose appears coherent, though scanners noted broad triggering, undeclared tool scope, and possible privacy exposure when content is sent to external services.
The verdict is benign, not risk-free, and confidence is medium. Before installing, consider whether broad activation is suitable for you and whether you may use it with sensitive content that could be sent to external services. The scan was partial: some referenced paths were unresolved, so the evidence is incomplete.
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
)
try:
with urllib.request.urlopen(req) as resp:
data = json.loads(resp.read().decode("utf-8"))
except urllib.error.HTTPError as e:
body = e.read().decode("utf-8", errors="replace")
Without declared permissions the skill's intent is opaque and cannot be validated.
The instruction explicitly activates the skill on “可视化” (visualization), which also covers ordinary charts, interactive dashboards, and other outputs outside infographic generation. No exclusion condition limits this trigger to requests for infographic-style images.
The workflow sends source-derived text to OpenRouter and a downstream image provider without an explicit sensitivity check or confirmation that external processing is authorized. Secret stripping reduces credential exposure but does not protect confidential business information, personal data, or other sensitive document content preserved verbatim.
The example recommends copying a source-derived prompt into external services without checking whether the document contains confidential or personal information. The manual handoff makes the external destination apparent and reduces automatic disclosure risk, but users may still expose protected verbatim content without understanding the provider's retention or processing policies.