The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requests access to the Bash tool in its configuration. In SKILL.md, it provides instructions for executing a Python script (scripts/generate_schematic.py) to generate diagrams from natural-language descriptions. This script is part of a referenced external skill (scientific-schematics).
[PROMPT_INJECTION]: The skill acts as a data ingestion point for untrusted user research information which is then processed into grant templates. There is a potential surface for indirect prompt injection as the skill lacks explicit instructions for the agent to use boundary markers or delimiters when handling user input, though this is considered a low-severity architectural risk in an educational context.
[EXTERNAL_DOWNLOADS]: The documentation in references/nstc_guidelines.md suggests that users download LaTeX templates and packages from public repositories, specifically GitHub and CTAN. These are well-known academic resources.

research-grants