The Agent Skills Directory

[COMMAND_EXECUTION]: The main script flexus_lifecycle.py executes a local helper script flexus_specs_extractor.py using subprocess.run. This is used to fetch updated specifications and is limited to executing code provided within the skill package.
[EXTERNAL_DOWNLOADS]: The script flexus_specs_extractor.py fetches configuration data from official Huawei Cloud documentation. This is a legitimate use of external resources for configuration updates.
[PROMPT_INJECTION]: The skill processes data from external documentation which creates a surface for indirect prompt injection. However, the risk is minimized by the structured parsing and validation of the data.
Ingestion points: scripts/flexus_specs_extractor.py fetches data from the vendor's documentation URL.
Boundary markers: None present.
Capability inventory: subprocess.run (to run internal scripts), requests.request (to interact with cloud APIs).
Sanitization: Content is parsed using regular expressions and validated against expected data types.
[SAFE]: The skill's instructions explicitly forbid the AI agent from revealing or logging credentials and recommend using environment variables for secret management.
[SAFE]: The scripts disable SSL certificate verification. While this is a security best practice violation, it does not indicate malicious intent and is often used for compatibility.

huawei-cloud-flexus-l-server-manage