The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided technical requirements to generate code. While this represents a standard ingestion surface for code generation agents, the skill includes explicit constraints that reinforce safety, such as mandatory Pydantic validation and type hinting, which mitigate accidental logic errors in the generated output.
Ingestion points: User-provided API requirements and data models in the Core Workflow section of SKILL.md.
Boundary markers: The instructions use structured sections and markdown blocks to separate requirements from implementation steps.
Capability inventory: The skill facilitates Python code generation for FastAPI, SQLAlchemy, and JWT authentication; it does not contain direct file-write or network-exfiltration logic.
Sanitization: The skill explicitly mandates the use of Pydantic V2 validation schemas for all data models to ensure runtime data integrity.
[COMMAND_EXECUTION]: The skill workflow recommends running pytest to verify implementation. This is a standard developer practice and is presented as a manual verification step for the user or agent environment rather than an automated, hidden execution of external scripts.
[CREDENTIALS_SAFE]: The skill demonstrates safe credential management by explicitly instructing the user to read the SECRET_KEY from environment variables rather than hardcoding it in the security implementation examples.

fastapi-expert