oraclecloud-prod-checklist
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using the OCI CLI and runs Python scripts using the OCI SDK to retrieve configuration data. This includes reading the OCI authentication file at
~/.oci/config, which is a standard requirement for the skill's intended purpose of infrastructure auditing. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection because it ingests and processes untrusted data from external cloud resource metadata.
- Ingestion points: Data is ingested via the output of
ociCLI commands and Python SDK calls (e.g., compartment lists, policy statements, security rule descriptions) as described inSKILL.md. - Boundary markers: The instructions do not define clear delimiters or include warnings for the agent to ignore potentially malicious instructions embedded within the resource data.
- Capability inventory: The skill has powerful capabilities across all scripts, including shell command execution (
Bash(oci:*),Bash(python3:*)) and file system operations (Read,Write,Edit). - Sanitization: There is no evidence of data sanitization, validation, or escaping of the metadata retrieved from the OCI environment before it is processed by the agent.
Audit Metadata