alibabacloud-sysom-diagnosis
Fail
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill uses
importlib.import_module()inscripts/sysom_cli/core/registry.pyto dynamically load command modules based on directory scanning at runtime. This dynamic loading pattern is used to register diagnostic commands from subdirectories. - [EXTERNAL_DOWNLOADS]: The skill fetches and executes scripts from remote URLs during installation and initialization. In
SKILL.md, it downloads an Aliyun CLI setup script fromhttps://aliyuncli.alicdn.com/setup.sh, and inscripts/init.sh, it downloads a package manager installer fromhttps://astral.sh/uv/install.sh. These downloads target the vendor's own infrastructure and a well-known service provider respectively. - [COMMAND_EXECUTION]: Multiple scripts execute system commands to collect diagnostic data.
scripts/sysom_cli/lib/kernel_log.pyexecutesjournalctlanddmesg, whilescripts/sysom_cli/memory/lib/classify_engine.pyexecutespsto sample process memory usage. - [CREDENTIALS_UNSAFE]: The skill interacts with local authentication configuration. In
scripts/sysom_cli/lib/auth.pyandscripts/sysom_cli/configure/command.py, the skill reads and writes to~/.aliyun/config.jsonand accesses environment variables likeALIBABA_CLOUD_ACCESS_KEY_IDto manage cloud credentials. It includes guidance to prevent users from sharing secrets directly in chat. - [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by processing untrusted external data.
- Ingestion points: System kernel logs collected in
scripts/sysom_cli/lib/kernel_log.pyand structured OpenAPI responses processed inscripts/sysom_cli/lib/diagnosis_helper.py. - Boundary markers: There are no explicit markers or delimiters used to separate untrusted system data from the agent's instructional context.
- Capability inventory: The skill has the ability to execute shell commands and perform network operations via the Aliyun SDK.
- Sanitization: No specific filtering or escaping logic was found for external data before it is presented to the agent.
Recommendations
- HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh, https://aliyuncli.alicdn.com/setup.sh - DO NOT USE without thorough review
Audit Metadata