The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill documents the use of SymPy's code generation and compilation utilities, specifically autowrap and ufuncify. These functions generate C or Fortran code from symbolic expressions and compile them into executable Python extensions at runtime to improve performance.
[DYNAMIC_EXECUTION]: The skill provides examples for using parse_expr, parse_latex, and parse_mathematica. These utilities dynamically interpret string input to create symbolic expression trees, which represents a form of dynamic logic generation.
[DYNAMIC_EXECUTION]: The reference materials include examples of using the pickle module to save and load SymPy objects. This is a standard serialization method, though the documentation notes that it should be handled with care regarding untrusted data.
[INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface where an agent might ingest untrusted string data and process it via mathematical parsers (e.g., Pattern 3: Interactive Computation). The documentation acknowledges this risk by explicitly instructing the user to validate and sanitize input to avoid code injection vulnerabilities.

sympy