The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill provides code patterns that read authentication data from ~/.oci/config. While this is a sensitive path containing credentials, its use is the standard and necessary method for authenticating the OCI SDK as described in the skill's primary purpose.
[EXTERNAL_DOWNLOADS]: Recommends the installation of the oci Python package. This is the official SDK for Oracle Cloud Infrastructure, which is a well-known service.
[INDIRECT_PROMPT_INJECTION]: The skill implements patterns for fetching and processing resource data from OCI (e.g., listing instances or regions), which introduces a surface for indirect prompt injection if the resource data contains malicious instructions.
Ingestion points: OCI SDK calls such as list_instances() and list_regions() in SKILL.md.
Boundary markers: None present in the provided code snippets to distinguish between external data and instructions.
Capability inventory: The skill is configured with Write, Edit, and Bash capabilities.
Sanitization: No explicit validation or sanitization of the remote resource metadata is demonstrated in the code examples.

oraclecloud-sdk-patterns