The Agent Skills Directory

[SAFE]: No prompt injection patterns or instructions to bypass safety guidelines were detected in the skill documentation or metadata.
[SAFE]: The skill does not contain hardcoded credentials, sensitive file path access, or unauthorized network operations.
[SAFE]: No obfuscation techniques, such as Base64 encoding of commands or hidden Unicode characters, were found.
[SAFE]: The skill does not download or execute remote scripts; all operations are conducted within the local analysis environment using built-in tools.
[COMMAND_EXECUTION]: The skill utilizes Bash and a SQL interface (idasql) to manage binary metadata and type information. These tools are used legitimately for the intended purpose of automating reverse engineering tasks.
[SAFE]: While the skill ingests metadata from the binary being analyzed (e.g., type and function names), which represents a surface for indirect prompt injection, it does not include instructions that would lead to unsafe execution of such data.

types