The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of LaTeX source files which can lead to indirect prompt injection attacks.
Ingestion points: The agent reads content from .tex files found in the working directory or subdirectories.
Boundary markers: Absent. The instructions do not specify any delimiters or warnings to the agent to ignore potentially malicious instructions embedded within the LaTeX source code.
Capability inventory: The skill performs file system read operations to locate and merge multiple LaTeX files and write operations to create the final translated Markdown document.
Sanitization: Absent. There are no instructions to sanitize the input text or validate the source content before it is processed for translation.

paper-translate