The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions specify access to sensitive credential files located at ~/.alibabacloud/credentials and sensitive environment variables (ALIBABACLOUD_ACCESS_KEY_ID, ALIBABACLOUD_ACCESS_KEY_SECRET, ALIBABACLOUD_REGION_ID). These files and variables contain secret keys used to authenticate with Alibaba Cloud services.
[EXTERNAL_DOWNLOADS]: The script scripts/list_openapi_meta_apis.py fetches OpenAPI metadata from Alibaba Cloud's official domain (api.aliyun.com) to discover available service APIs and schemas.
[INDIRECT_PROMPT_INJECTION]: The skill processes external JSON data retrieved from the Alibaba Cloud API metadata service, which could potentially contain malicious instructions.
Ingestion points: The fetch_json function in scripts/list_openapi_meta_apis.py retrieves content from a remote URL.
Boundary markers: The skill instructions do not include delimiters or specific warnings to ignore instructions embedded within the retrieved metadata.
Capability inventory: The skill has the ability to execute Python scripts (scripts/list_openapi_meta_apis.py), write files to the local output directory, and perform network operations.
Sanitization: No sanitization, schema validation, or filtering is performed on the retrieved JSON data before it is parsed and written to local markdown and JSON files.
[COMMAND_EXECUTION]: The skill includes instructions to execute a local Python script (scripts/list_openapi_meta_apis.py) and utilizes the py_compile module to validate the syntax of the skill's own script files.

aliyun-gbi-analytics