nodel-dev
Warn
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the Gradle wrapper (
./gradlew) for building, running, and testing the platform. - [EXTERNAL_DOWNLOADS]: The build pipeline for
nodel-webui-jsdownloads Node.js and npm packages from public registries. Additionally, the Playwright testing setup triggers the download of the Chromium browser via./gradlew :nodel-jyhost:playwrightInstall. - [REMOTE_CODE_EXECUTION]: The platform features an embedded Jython 2.5.4-rc1 environment (
nodel-jyhost) that executes scripts for individual nodes. Furthermore, the discovery mechanism (AutoDNS.java) allows for dynamic class loading based on a system property (org.nodel.discovery.impl), which can load and execute arbitrary Java classes provided on the classpath. - [PROMPT_INJECTION]: The skill manages a platform that processes untrusted data via line-delimited JSON over TCP and REST API calls. While it includes name reduction and normalization, the direct execution of scripts based on these inputs presents a surface for indirect prompt injection (Category 8).
Audit Metadata