init
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes arbitrary shell commands provided through the
--evalparameter. This is a core feature used to measure baseline metrics for the tasks agents are intended to perform. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the
--taskand--evalarguments which are then used in command execution and session configuration. If an agent populates these arguments based on malicious instructions found in external data (like a README or a website), it could lead to unintended command execution. - Ingestion points: The
SKILL.mdfile defines parameters for--taskand--evalthat are passed to a Python initialization script. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill definition.
- Capability inventory: The skill possesses the ability to execute shell commands in the current working directory to capture baselines.
- Sanitization: There is no evidence of input validation, escaping, or filtering for the strings provided in the command arguments.
Audit Metadata