The Agent Skills Directory

[SAFE]: The skill is entirely composed of documentation and templates for systems thinking protocols (Soft Systems Methodology, Critical Systems Heuristics, and System Dynamics). It lacks any functionality for code execution, file system modification, or network communication.
[PROMPT_INJECTION]: The skill is designed to process user-supplied descriptions of complex problems. 1. Ingestion points: User input is ingested via descriptions in Stage 1 and Stage 2. 2. Boundary markers: No explicit markers (e.g., XML tags) are used to wrap user input. 3. Capability inventory: The skill has no access to tools, subprocesses, or file-writing functions. 4. Sanitization: None. Because the skill has no dangerous capabilities, the lack of sanitization or delimiters does not present an exploitable vulnerability surface.

systems-practice