The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to utilize the Bash tool to execute internal Python scripts such as generate_schematic.py, compliance_checker.py, budget_calculator.py, and deadline_tracker.py. These scripts are intended for task-specific functions like figure generation and budget calculations. The use of these scripts is a standard feature for this vendor's productivity ecosystem.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) related to its schematic generation feature. User-provided descriptions are passed as command-line arguments to a shell script, which could be exploited if the input is not correctly handled by the agent or the script.
Ingestion points: User natural language descriptions for diagrams in the SKILL.md and references/README.md files.
Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are present in the provided command templates.
Capability inventory: The skill allows the use of the Bash tool to run Python scripts with user-supplied arguments.
Sanitization: No sanitization or escaping logic is visible in the markdown instructions for handling user input before shell execution.
[EXTERNAL_DOWNLOADS]: The documentation references official government domains (e.g., nsf.gov, nih.gov, energy.gov, darpa.mil) and community-maintained LaTeX templates on GitHub. These are recognized as legitimate informational resources and do not involve the dynamic loading or execution of remote code by the skill itself.

research-grants