The Agent Skills Directory

[SAFE]: The skill serves as a technical manual for security researchers, providing guidance on analyzing embedded firmware running various RTOS kernels. It contains no executable malicious logic or instructions that attempt to bypass AI safety guidelines.
[EXTERNAL_DOWNLOADS]: The documentation references official tools such as Ghidra (via the National Security Agency's GitHub repository) and binwalk (via standard package managers and PyPI). These are well-known, trusted resources used for firmware analysis.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external firmware data. While this represents a standard analysis attack surface, the documentation provides best practices for static analysis and does not instruct the agent to execute untrusted code found within the firmware.
[DYNAMIC_EXECUTION]: Mentions using Ghidra scripts and Python to automate the parsing of task tables. These are standard, legitimate activities for a reverse engineering tool and are not used here for malicious purposes.

re-rtos