code-engine-specialist-skill
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
ibmcloud ceCLI to manage cloud resources. This behavior is consistent with its stated purpose as a Code Engine specialist and is controlled via a workflow that requires user confirmation before executing deployment commands. - [DATA_EXPOSURE]: The skill analyzes local project metadata (e.g.,
package.json,Dockerfile,.env.example) to determine deployment requirements. It correctly advises users to store sensitive credentials in Code Engine secrets rather than hardcoding them in source code or configuration files. - [EXTERNAL_DOWNLOADS]: The skill references container base images from well-known and trusted registries, including the Red Hat Ecosystem (registry.access.redhat.com) and Google Container Registry (gcr.io). It also supports building from user-provided Git repositories, which is a standard feature for cloud deployment tools.
- [SAFE]: (Indirect Prompt Injection Surface Analysis)
- Ingestion points: Phase 1 reads local project files such as
package.json,requirements.txt, andDockerfileto extract configuration details. - Boundary markers: No explicit markers are used, but the skill interprets the data for keyword extraction and resource planning rather than direct prompt interpolation.
- Capability inventory: The skill can execute
ibmcloud cecommands and read/list files. - Sanitization: The workflow includes Phase 3 (Interactive Planning) which requires the user to review recommendations and confirm choices before any execution occurs in Phase 5.
Audit Metadata