The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data (existing C++ header files) to generate documentation. While this creates a potential surface for indirect prompt injection if a header file contains malicious instructions, the skill is scoped strictly to adding comments.
Ingestion points: C++ header files (*.hpp, *.h) located in src/ and include/ directories.
Boundary markers: Absent; the skill does not explicitly instruct the agent to ignore instructions embedded within the code it processes.
Capability inventory: File modification (writing documentation comments to existing header files).
Sanitization: Absent; the skill relies on the agent's internal logic to distinguish between code and documentation tasks.
[EXTERNAL_DOWNLOADS]: The skill references official documentation from trusted sources such as Doxygen and the Google C++ Style Guide for educational purposes. These references do not involve automated downloads or execution of remote code.
[COMMAND_EXECUTION]: The skill explicitly instructs the agent to avoid creating new resource files, scripts, or configuration files (like Doxyfile), which mitigates risks associated with unintended command execution or environment modification.

cpp-api-documentation