The Agent Skills Directory

[COMMAND_EXECUTION]: The skill orchestrates debugging sessions by executing system-level debuggers including gdb, lldb, and debugpy. It requires access to process memory and the local filesystem to inspect variables and manage breakpoints.
[EXTERNAL_DOWNLOADS]: Documentation instructs the user to install external dependencies such as the dap-mcp server and language-specific debuggers from standard package managers like pip and npm.
[PROMPT_INJECTION]: An indirect prompt injection surface exists because the skill reads and processes local source code files and project metadata. Ingestion occurs when the agent loads file content for debugging. There are no explicit boundary markers or sanitization routines defined to mitigate malicious instructions embedded in the code being debugged. The skill operates with high-privilege capabilities including filesystem manipulation and subprocess control.

dynamic-debugger