The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to utilize the Bash tool to execute the xelatex command. This is used for compiling LaTeX source files into professional reports, which is a legitimate functional capability for the stated purpose of the skill.
[PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection due to the way it processes external data.
Ingestion points: Untrusted data enters the agent's context through the research-lookup tool during the first stage of the manuscript development workflow (Stage 1: Planning).
Boundary markers: The skill does not instruct the agent to use boundary markers or delimiters when handling research data, nor are there instructions to ignore instructions embedded within that data.
Capability inventory: The skill is configured with access to the Write, Edit, and Bash tools, providing it with significant capabilities to modify the local environment based on processed content.
Sanitization: There is no explicit requirement or process defined for sanitizing or validating the content retrieved from external lookups before it is used to generate manuscript paragraphs.

scientific-writing