penetration-testing-with-strix
Fail
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides a command to install the Strix CLI by piping a remote script to the shell:
curl -sSL https://strix.ai/install | bash. This executes code from the vendor's infrastructure directly on the local machine. - [EXTERNAL_DOWNLOADS]: The skill downloads installation scripts from
strix.aiand interacts with the managed cloud service atapp.strix.aifor authenticated scanning and reporting. - [COMMAND_EXECUTION]: The skill invokes the
strixCLI tool to perform various security audits, including local white-box scans where code directories are mounted into a sandbox environment. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The agent ingests data from external targets including deployed URLs, remote GitHub repositories, and OpenAPI/Postman specifications provided via the
-tor--targetflags inSKILL.md. - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from interpreting malicious instructions embedded within the scanned target's code or responses.
- Capability inventory: The tool possesses high-privilege capabilities including network exploitation, file system access (mounting local paths), and the ability to interact with LLMs to generate and prove exploits.
- Sanitization: No specific sanitization or filtering logic is mentioned for the content retrieved from the targets before it is processed by the AI agents.
Recommendations
- HIGH: Downloads and executes remote code from: https://strix.ai/install - DO NOT USE without thorough review
Audit Metadata