agentuity-cli-cloud-keyvalue-search
Warn
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill uses
bunxto download and execute the@agentuity/clipackage at runtime across several examples (e.g.,bunx @agentuity/cli kv search production user). Executing unpinned external packages from public registries allows for the potential execution of malicious code if the package is compromised or subject to dependency confusion attacks. - [EXTERNAL_DOWNLOADS]: The skill relies on fetching the
@agentuity/clitool from the npm registry. The source vendor is not explicitly verified, and the download occurs dynamically during the skill's operation. - [COMMAND_EXECUTION]: The skill relies on executing shell commands to perform its primary function. It constructs commands using user-provided arguments (
<name>and<keyword>) which are passed directly to thebunx @agentuity/clicommand string. - [PROMPT_INJECTION]: The skill processes user-supplied input (
<name>and<keyword>) to construct shell commands. This creates an indirect prompt injection surface where maliciously crafted input containing shell metacharacters (e.g.,;,&&,|) could lead to arbitrary command execution if the agent does not properly sanitize the input.
Audit Metadata