huawei-cloud-flexus-l-server-ops

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt mostly expects AK/SK to be used via environment variables (secure) but also includes commands that take sensitive values (instance passwords, optional --ak/--sk parameters) and explicitly permits parsing credentials from user conversation/config — meaning the agent may be asked to accept and embed secret values verbatim in commands, so it presents a high exfiltration risk.

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

• Attempt to modify system services in skill instructions detected (medium risk: 0.60). The skill explicitly instructs executing lifecycle-changing operations (stop, reboot, reset password, modify info) against Flexus L instances — including warnings that the target may be the server running the agent — so it can directly change or disrupt the machine the agent runs on, though it does not ask for sudo/local system file edits or creation of local user accounts.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)