check-plain-env
Fail
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill's primary function is to execute shell commands to probe the host environment for toolchains, service binaries, and hardware accelerators.
- [EXTERNAL_DOWNLOADS]: The skill references remote installation scripts, specifically the official Rust installer (
https://sh.rustup.rs). These are documented as suggestions for the user and are associated with a well-known service. - [COMMAND_EXECUTION]: There is a potential for command injection via indirect prompt injection. The agent generates the list of commands to be executed by parsing untrusted repository content, such as filenames and code within
.plainand script files. Maliciously crafted content could attempt to inject shell operators into the version-checking commands. - [SAFE]: The skill contains explicit safeguards against credential exposure. It instructs the agent to verify the existence of environment variables (e.g.,
CODEPLAIN_API_KEY) using methods that avoid printing their values to the output. - [SAFE]: The instructions establish a strict 'read-only' policy, explicitly preventing the agent from performing installations or system modifications without direct user intervention.
Recommendations
- HIGH: Downloads and executes remote code from: https://sh.rustup.rs - DO NOT USE without thorough review
Audit Metadata