The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the Bash tool to execute python scripts/generate_schematic.py. This script is not included in the provided 7 files, making its logic and safety unverifiable during analysis.- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by processing untrusted scientific claims and research papers and then using natural language descriptions from those sources as arguments for a shell command. Mandatory Evidence: 1) Ingestion points: Reads external scientific claims and research papers via user input or file reads in SKILL.md; 2) Boundary markers: No delimiters or instructions to ignore embedded commands are present when processing external text; 3) Capability inventory: The skill has access to Bash, Write, Edit, and Read tools; 4) Sanitization: There is no evidence of sanitization or escaping of the input text before it is passed to the generate_schematic.py script as a command-line argument.

scientific-critical-thinking