alibabacloud-workbench-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the Agent to write credentials directly into ~/.workbench/config.json (including access_key_id and access_key_secret), which requires the LLM to output secret values verbatim into commands/files.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). These are direct remote install scripts (install.sh / install.ps1) hosted on an OSS object-storage domain and invoked by piping to a shell/iex — a high-risk pattern because remote scripts can contain malicious payloads even if hosted on cloud storage.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill's install steps fetch and execute remote scripts at runtime (curl ... | bash and irm ... | iex), specifically https://workbench-cli.oss-cn-hangzhou.aliyuncs.com/install.sh and https://workbench-cli.oss-cn-hangzhou.aliyuncs.com/install.ps1, so remote code controls execution.