huawei-cloud-codearts-code-gen

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the agent to request the user's AK/SK, store them in plaintext (~/.bashrc), and execute commands embedding the credentials verbatim (e.g., setup.sh --save-aksk "user's AK" "user's SK"), which forces the LLM to handle/output secret values directly.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The package contains deliberate high-risk behaviors: it forcefully mandates running a setup that downloads and executes remote install scripts, persistently stores cloud AK/SK in plaintext (or user environment), and writes broad file-permission policies that enable wide read/write access — patterns that enable supply-chain execution, credential capture, and potential filesystem compromise.

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

• Potentially malicious external URL detected (high risk: 1.00). The setup scripts fetch and execute remote install scripts at runtime from https://cnnorth4-cloudide-marketplace.obs.cn-north-4.myhuaweicloud.com/codearts/cli_tui/install_script/install.sh and https://cnnorth4-cloudide-marketplace.obs.cn-north-4.myhuaweicloud.com/codearts/cli_tui/install_script/install.ps1, which are downloaded during runtime and executed (so remote code controls execution and is required for the skill).

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). This skill explicitly instructs the agent to auto-download and install a CLI, modify the user's shell config (~/.bashrc) to store plaintext AK/SK, and run permission scripts that create/modify files in the workspace—actions that change the machine's state and handle sensitive credentials.