hardware-technique
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of system utilities to interact with hardware and network targets. This includes
nmapfor discovery,ncfor raw socket communication,curlfor web management interfaces, and specialized hardware tools likeflashrom,openocd,minicom, andscreenfor console and memory access. - [EXTERNAL_DOWNLOADS]: The skill references and provides commands to download external tools from public repositories, specifically the Printer Exploitation Toolkit (PRET) from
github.com/RUB-NDS/PRET. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent processes data from potentially untrusted external sources, such as firmware images, serial console boot logs, and printer protocol responses.
- Ingestion points: Network responses from TCP 9100, serial data from
/dev/ttyUSB0, and extracted filesystem artifacts from firmware images. - Boundary markers: The instructions do not define specific delimiters for separating target data from agent instructions during analysis.
- Capability inventory: The agent is granted capabilities to execute shell commands, perform network requests, and write to the local filesystem via firmware extraction tools.
- Sanitization: No specific sanitization steps are documented for handling data retrieved from hardware targets before it is analyzed by the agent.
Audit Metadata