dsh-versatileds-plugin@0.3.1
AI Security Review
scanned 10d ago · by lpm-firewall-aiThe source supports a conditional credential-exposure warning because a separately sourced DeepSeek key can be sent to an unrestricted configured endpoint. The default is service-aligned and no package-controlled recipient or malicious intent was established.
Decision evidence
public snapshot- The QQ model key is loaded from configuration, the environment, or a local key file.
- The QQ model endpoint can be configured without a host restriction; its default is DeepSeek.
- Model requests send the loaded key as a bearer credential along with conversation content.
- Model calls occur through QQ proactive and poke reply paths when a key and endpoint are available.
- Bili model requests use the DSH llm stream with a configurable provider and model, sending the prompt as user-role text.
- The configured default endpoint is DeepSeek, consistent with the named DeepSeek API key.
- The source does not establish a fixed package-controlled recipient; exposure depends on the configured endpoint.
Source & flagged code
5 flaggedFindings
Affected versions and remediation
This report applies to dsh-versatileds-plugin@0.3.1.
See all versions of dsh-versatileds-plugin for other recorded verdicts.
- Review the evidence and your use of dsh-versatileds-plugin@0.3.1 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.