oraclecloud-core-workflow-a

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses local configuration files to authenticate with Oracle Cloud Infrastructure. This is the standard method for using the OCI SDK to manage cloud resources.
  • Sensitive file paths accessed: ~/.oci/config (OCI credentials) and /home/user/.ssh/id_rsa.pub (SSH public key).
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates a workflow that ingests content from a local file and incorporates it into instance metadata, creating a vulnerability surface for indirect instructions.
  • Ingestion points: The skill reads ~/.ssh/id_rsa.pub to provide SSH access to new instances.
  • Boundary markers: None present in the provided code snippets.
  • Capability inventory: The skill uses the oci library to perform instance launch, lifecycle actions (start/stop), and network configuration.
  • Sanitization: No explicit sanitization or validation of the public key file content is shown before it is sent to the OCI API.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:46 AM
Security Audit — agent-trust-hub — oraclecloud-core-workflow-a