The Agent Skills Directory

[SAFE]: The skill contains documentation and Jython scripts for Ghidra-based firmware analysis. All scripts operate locally within the Ghidra environment and do not perform network operations or access sensitive system files outside of the project scope.
[COMMAND_EXECUTION]: The skill provides instructions for using Ghidra's analyzeHeadless utility to automate binary analysis. These are standard developer operations for the tool described.
[INDIRECT_PROMPT_INJECTION]: As a reverse engineering tool, this skill is designed to process untrusted binary data. The analysis is static, and the scripts do not execute the target binary's code or interpret it as instructions for the agent.
Ingestion points: Binary files imported into Ghidra via analyzeHeadless (documented in SKILL.md).
Boundary markers: Not applicable for binary code analysis.
Capability inventory: The skill uses local Jython scripts to interact with the Ghidra API for renaming functions, adding comments, and labeling data.
Sanitization: Not applicable as the skill performs static analysis rather than code execution of the target.

ghidra-re