@luoyi-dsh/discord-plugin@1.0.4
DeepSeek Harness (DSH) Discord 用戶帳號 (Selfbot) 助理與 MCP 插件:支援自然穿插閒聊、被提及與私訊智慧代回、傳送文字、圖片、文件與貼圖
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe flagged LLM code uses credentials for requests to their configured or corresponding provider services, and the prepare script only builds the package. The automatic Discord messaging features match the package’s stated purpose and are controlled through its configuration.
Decision evidence
public snapshot- The only lifecycle script is prepare, which runs the TypeScript build.
- LLM API keys are resolved from plugin configuration, provider-specific environment variables, or DSH credentials.
- Google OAuth refresh tokens are sent to Google’s OAuth endpoint; subscription access tokens go to Google’s Cloud Code endpoints.
- Custom API keys are sent as bearer credentials to the configured API base, defaulting to api.deepseek.com.
- The package entrypoint loads configuration and creates a Discord client; login requires a configured user token and autoStart.
- Discord replies and autonomous chat are advertised package features, with controls and limits in configuration.
- The LLM code reads API keys from environment variables and a local credentials file, and reads subscription tokens from local auth.json.
- Default settings enable mention replies, DM replies, and autonomous co-chat when a Discord token is configured.
Source & flagged code
4 flaggedPackage source references dynamic require/import behavior.
client.jsView on unpkg · L18Source appears to send environment or credential material to an external endpoint.
dist/lib/llm.jsView on unpkg · L17Findings
Affected versions and remediation
This report applies to @luoyi-dsh/discord-plugin@1.0.4.
See all versions of @luoyi-dsh/discord-plugin for other recorded verdicts.
- Review the evidence and your use of @luoyi-dsh/discord-plugin@1.0.4 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.