rhdh-local

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes container orchestration commands (podman/docker compose) and git status via subprocess calls. These operations are core to the skill's purpose of managing a local development environment and are invoked using fixed command structures.
  • [EXTERNAL_DOWNLOADS]: The fetch-plugin-metadata.py script retrieves plugin configuration examples and OCI artifact metadata from the official redhat-developer GitHub repository. These requests target a trusted vendor source and do not involve direct execution of downloaded content.
  • [PROMPT_INJECTION]: The skill processes external plugin metadata which serves as a potential ingestion point for indirect prompt injection. The risk is mitigated by the skill's instructions to use references verbatim and the restriction against reading secret values into the model context.
  • [SAFE]: The backup and restore functionality in rhdh_local/backup.py includes robust path validation that prevents path traversal attacks by ensuring restored files remain within the intended workspace directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 03:41 PM
Security Audit — agent-trust-hub — rhdh-local