The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the pydantic-settings library, which is a widely recognized and trusted extension for managing application configuration via environment variables.\n- [PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection as it is designed to parse and validate untrusted data from external sources such as APIs and user-provided JSON.\n
Ingestion points: Data enters through model_validate, model_validate_json, and BaseSettings interfaces in SKILL.md and templates/model_patterns.py.\n
Boundary markers: The models use strict schemas with constraints like min_length, regex pattern matching, and extra="forbid" to establish clear data expectations.\n
Capability inventory: The skill is limited to data validation and does not perform high-risk actions like subprocess execution or file system modifications with the validated data.\n
Sanitization: Inputs are sanitized using Pydantic's type-enforcement system and custom logic in field_validator and model_validator methods to ensure data integrity.\n- [SAFE]: No malicious patterns, hidden commands, or unauthorized data exfiltration attempts were found in the skill code or documentation.

pydantic-models