rhdh-upgrade-helper
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches productized plugin metadata and documentation from trusted sources, including 'quay.io/rhdh/plugin-catalog-index', 'docs.redhat.com', and the official 'redhat-developer/rhdh-plugin-export-overlays' GitHub repository.
- [COMMAND_EXECUTION]: Utilizes standard developer tools such as
oc,podman,docker,git, and theghCLI to extract image data, clone repositories, and search for community-reported bugs. These operations are restricted to the vendor's known infrastructure and well-known services. - [DATA_EXFILTRATION]: Reads user configuration files and environment files (
.env) to analyze the current setup. This involves accessing potentially sensitive data; however, the skill includes a mandatory pre-processing step defined inreferences/secrets-detection.mdthat scans for API keys and passwords, warns the user, and redacts identified values from the final report. - [INDIRECT_PROMPT_INJECTION]: Processes external content from official documentation and bug trackers (Jira and GitHub). While this data is technically untrusted, the ingestion is performed for the primary purpose of summarizing release notes and known issues, with the output structured into predefined markdown sections.
Audit Metadata