mthds-run
Fail
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill contains instructions for users to install the required Pipelex runtime via 'curl -fsSL https://pipelex.com/install.sh | sh'. This is a high-risk execution pattern that allows arbitrary code from an unverified remote server to be executed with the user's local shell privileges.
- [COMMAND_EXECUTION]: The skill executes multiple shell commands using 'mthds-agent' and 'pipelex-agent' for version checks, configuration diagnostics, and the execution of pipeline bundles.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes external data.
- Ingestion points: The agent reads and processes data from 'inputs.json' and the output of CLI commands.
- Boundary markers: There are no boundary markers or instructions to isolate the untrusted data from the agent's core instructions.
- Capability inventory: The skill allows for command execution and file reading via the MTHDS CLI.
- Sanitization: No validation or sanitization of the input JSON or tool output is performed before the agent acts on or presents the data.
Recommendations
- HIGH: Downloads and executes remote code from: https://pipelex.com/install.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata