limrun-gradle
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides the
lim gradle runcommand, enabling the execution of arbitrary shell commands within the remote Gradle sandbox. - [EXTERNAL_DOWNLOADS]: Users are instructed to install the
limCLI tool globally via npm (npm install --global lim) to interface with the vendor's remote services. - [DATA_EXFILTRATION]: The skill synchronizes the contents of the local project directory to the vendor's infrastructure at
limrun.comto facilitate remote builds. - [INDIRECT_PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting and processing untrusted data from local project files. \n- Ingestion points: The
lim gradle build .command synchronizes the entire current directory to the remote environment. \n- Boundary markers: None identified in the instructions to delimit or ignore instructions within synced files. \n- Capability inventory: Includes remote shell execution (lim gradle run), Gradle task execution, and environment configuration. \n- Sanitization: No explicit sanitization or validation of the synced file contents is described. - [DYNAMIC_EXECUTION]: The
lim gradle useandlim gradle toolscommands allow for the dynamic selection and installation of toolchains such as Node.js, Java, and pnpm at runtime within the sandbox environment.
Audit Metadata