dependency-audit
Fail
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: HIGHPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill uses
sudocommands to install thetrivyvulnerability scanner and its repository keys. This involves executingsudo apt-get install,sudo apt-key add, andsudo teeto modify system-level package manager configurations. - [EXTERNAL_DOWNLOADS]: The skill fetches the official repository key and package information for
trivyfrom Aqua Security's infrastructure (aquasecurity.github.io). - [COMMAND_EXECUTION]: The skill executes various shell commands to detect project types and run audit tools. These include
npm install --package-lock-only,npm audit,pip-audit,trivy fs, and severaljqoperations to parse audit results. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted project configuration files, creating a surface for potential instruction injection.
- Ingestion points: Project manifest files including
package.json,package-lock.json,yarn.lock,requirements.txt,Pipfile,pyproject.toml,go.mod, andDockerfilefrom the path specified inSECSKILL_PROJECT_PATH. - Boundary markers: None. The content of these files is read and passed directly to audit tools or processed via shell commands without delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill has capabilities for file system writes (
mkdir, output redirection), file system navigation (cd), and execution of multiple CLI tools. - Sanitization: There is no explicit sanitization of the data read from project files before it is processed by
jqor included in the final summary report generated for the agent.
Recommendations
- AI detected serious security threats
Audit Metadata