claude-to-im

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the agent to collect API tokens/secret credentials from the user and to write them into ~/.claude-to-im/config.env and use them for validation/daemon commands, which requires the model to handle and embed secret values verbatim (even if masked in UI), creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill directly ingests and forwards untrusted user-generated messages from public IM platforms (Telegram, Discord, Feishu/Lark, QQ, WeChat, DingTalk) to the Claude/Codex agent as part of its core workflow (see README.md and SKILL.md "Messages from IM are forwarded to the AI coding agent" and the setup/handoff/permission flows), so third-party content can influence agent decisions and tool use.