Security audit
Security checks for vulnerabilities and agentic risk
This is a simple weather skill that calls wttr.in for forecasts and does not show hidden persistence, credential access, destructive behavior, or unrelated data collection.
Before installing, understand that weather requests are sent to wttr.in, and if you ask without specifying a city the service may infer your approximate location from network/IP information. Use an explicit city if you want to avoid automatic location lookup.
The skill declares executable behavior via a shell-invoked Python script and requires external binaries, but it does not declare any explicit tool scope or permissions. This weakens least-privilege controls and makes it harder for a host platform to constrain or audit what commands the skill is allowed to run, increasing the risk of unintended command execution or capability creep.
The skill advertises IP-based automatic location lookup but does not warn users that invoking it without a city may disclose approximate location to an external weather service. This is a privacy issue because users may not realize that a seemingly simple weather query can trigger geolocation by IP and outbound requests to a third party.
The trigger phrases are broad enough to match ordinary conversation, which can cause the skill to activate unexpectedly without clear user intent. In a skill that may perform network access and possibly IP-based location lookup, accidental activation can lead to unnecessary data exposure or surprising behavior even if the functionality is not inherently high risk.
The skill sends either a user-provided city or, when no city is supplied, an IP-based location query to wttr.in without any disclosure or consent flow. This can leak user location or inferred location metadata to a third-party service, which is especially relevant because the default behavior performs IP-based lookup automatically.
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
try:
# 使用 format=j1 获取 JSON 格式数据
cmd = ["curl", "-s", f"wttr.in/{city}?format=j1"]
result = subprocess.run(cmd, capture_output=True, text=True, timeout=10)
if result.returncode != 0:
return {"error": "网络请求失败"}
data = json.loads(result.stdout)
The natural-language docstring is written in Chinese, and the script's user-facing output strings are also hard-coded in Chinese. This imposes a specific language on users without any opt-in or documented locale constraint.
No suspicious patterns detected.