The Agent Skills Directory

[SAFE]: The skill is entirely focused on providing system design fundamentals and does not contain any malicious instructions or safety filter bypass attempts.
[COMMAND_EXECUTION]: The inclusion of scripts/validate.py is a standard developer utility for ensuring the skill directory follows the expected structure. The script does not utilize eval(), exec(), or subprocess calls for arbitrary command execution.
[DATA_EXFILTRATION]: Analysis of the skill and its accompanying scripts confirmed no network operations (curl, wget, fetch) or attempts to access sensitive system files (e.g., .ssh, .aws, .env).
[REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. All logic is contained within the provided local files, and dependencies are limited to standard libraries and verified packages like PyYAML.
[PROMPT_INJECTION]: The SKILL.md file uses structured YAML schemas for parameter validation and does not include instructions to ignore previous constraints or bypass agent safety guardrails.
[INDIRECT_PROMPT_INJECTION]: The skill defines an ingestion point for external data via the design_context parameter. However, the risk is negligible as the skill lacks side-effect capabilities such as file system writing or network communication.
Ingestion points: design_context parameter in SKILL.md.
Boundary markers: The skill utilizes a strict YAML schema to validate the structure and length of the input problem statement.
Capability inventory: No scripts or tools perform network operations, file writes, or command execution.
Sanitization: Input is validated against length and type constraints before processing.

system-design-basics