gitlab-ci-job-configuration
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- Prompt Injection (SAFE): No patterns of instruction override, jailbreak attempts, or system prompt extraction were found in the markdown or metadata.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths, or unauthorized network operations are present. All examples use placeholder URLs and standard environment variables.
- Unverifiable Dependencies & Remote Code Execution (SAFE): The skill does not download or execute remote scripts. Packages mentioned (like npm) are illustrative examples within static code blocks.
- Privilege Escalation (SAFE): There are no commands targeting root access or system-level privilege changes.
- Obfuscation (SAFE): No encoded strings, homoglyphs, or hidden characters were detected in the content.
- Indirect Prompt Injection (SAFE): The skill provides static templates for CI/CD pipelines. While the agent possesses tools like Bash and Write, the skill itself does not ingest or process untrusted external data in a way that creates an injection surface.
Audit Metadata