jfrog-devops
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes official vendor resources including the JFrog CLI and specific GitHub Actions (
jfrog/setup-jfrog-cli,jfrog/frogbot) to perform artifact management and security scanning tasks. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from Artifactory which could potentially contain malicious instructions embedded by an external party in artifact metadata or repository content.
- Ingestion points: The agent ingests artifact properties, Artifactory Query Language (AQL) search results, and repository metadata via
scripts/artifactory_client.pyandscripts/xray_client.py. - Boundary markers: No specific delimiters or boundary warnings are implemented to distinguish ingested metadata from instructional prompts.
- Capability inventory: The agent has the capability to write to the local file system (artifact downloads), deploy artifacts to remote repositories, and execute JFrog CLI commands (
jf). - Sanitization: There is no evidence of explicit sanitization, filtering, or escaping of external artifact properties or AQL result values before they are processed by the agent.
Audit Metadata