ddev-development
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for configuring
hookswithexec-host, which triggers arbitrary command execution on the host system when DDEV project events occur (e.g.,post-start). It also documentsweb_extra_daemonswhich run background commands within the container. - [REMOTE_CODE_EXECUTION]: The documentation includes the
ddev add-on get <repo>command, which fetches and executes code and configuration from remote third-party GitHub repositories (e.g.,codingsasi/ddev-playwright). - [INDIRECT_PROMPT_INJECTION]: The skill involves reading and acting upon project-specific configuration files such as
.ddev/config.yamland.ddev/web-build/Dockerfile. These files serve as ingestion points for untrusted data that can trigger powerful capabilities like command execution and remote downloads without specific boundary markers or sanitization checks mentioned for the agent's interaction.
Audit Metadata