ide-management
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines several shell commands using the
acli(Acquia CLI) tool to perform operations on Cloud IDE environments. These tasks include listing (acli ide:list), creating (acli ide:create), opening (acli ide:open), and deleting IDEs, as well as configuring system services like Apache and PHP. This represents the primary intended functionality of the skill for administrative users.\n- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection where untrusted external data could be processed via shell commands.\n - Ingestion points: User-controlled parameters such as
<applicationUuid>,<version>, and IDE labels specified in theSKILL.mdinstructions.\n - Boundary markers: None identified; user variables are passed directly as arguments to the CLI tool.\n
- Capability inventory: Full shell execution capabilities through the provided
aclicommand set.\n - Sanitization: The skill instructions do not specify sanitization or validation logic, relying on the underlying
aclibinary for input handling.
Audit Metadata