The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes security scan findings from an external API, creating a potential attack surface for indirect prompt injection if malicious content is embedded in the scanned files' metadata or results.
Ingestion points: The agent interprets the check-report.json file generated by the scanning script.
Boundary markers: Instructions do not specify explicit delimiters or guardrails to differentiate scan data from agent instructions.
Capability inventory: The skill has the ability to execute local scripts, read and write files, and perform network requests via the AISC SDK.
Sanitization: No specific sanitization process is described for the scan report content before it is read by the agent.
[DYNAMIC_EXECUTION]: The detection script utilizes dynamic importing of libraries.
Evidence: scripts/skill_file_check.py includes a _load_sdk() function that imports the Alibaba Cloud SDK modules at runtime.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of external Python dependencies from a trusted vendor.
Evidence: scripts/requirements.txt lists several Alibaba Cloud SDK packages (alibabacloud_aisc20260101, alibabacloud_credentials, alibabacloud_tea_openapi) necessary for API interaction.

alibabacloud-aisc-skill-inspection