Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward Lark/Feishu bot usage guide whose messaging, upload/download, and listener behavior matches its stated purpose.
Install only if you intend to let the agent operate a Lark/Feishu bot. Use least-privilege app credentials, restrict the bot to intended chats, review files before upload, and store downloaded attachments in temporary or controlled directories with cleanup after processing.
Data is uploaded to cloud storage (S3 / GCS / Azure Blob). This may be a legitimate backup or exfiltration to an external bucket. Manual review is recommended.
# 上传文件
file_key = bot.upload_file("/path/to/file.pdf", file_type="pdf")
# 发送音频
bot.send_audio_to_user(user_open_id, file_key)
Data is uploaded to cloud storage (S3 / GCS / Azure Blob). This may be a legitimate backup or exfiltration to an external bucket. Manual review is recommended.
# 上传文件
file_key = bot.upload_file("/path/to/file.pdf", file_type="pdf")
# 发送音频
bot.send_audio_to_user(user_open_id, file_key)
The skill explicitly documents downloading images, files, and message resources to caller-specified local paths, but it does not warn that this writes potentially sensitive remote content onto local storage where it may persist beyond the intended task. In an agent setting, that omission can lead to unreviewed retention of confidential data, accidental re-use, or unsafe handling of untrusted files.
The listener section states that incoming images and files are downloaded to a temporary directory for processing, but it does not foreground the privacy and security implications of automatically materializing user-supplied content on local storage. This is risky because inbound files may contain sensitive information or malicious payloads, and users of the skill may not realize local persistence is occurring during message handling.
No suspicious patterns detected.