The Agent Skills Directory

[SAFE]: The skill is entirely composed of Markdown documentation and illustrative Clojure code snippets. It contains no executable scripts, remote downloads, or instructions that interact with the host system, network, or sensitive files.
[PROMPT_INJECTION]: As a data validation skill, it documents ingestion points for untrusted data (e.g., API requests and form validation) using functions like m/validate and m/coerce. While this represents a surface for indirect prompt injection, it is the intended use case for the library and Malli provides robust data-level sanitization capabilities. There are no instructions attempting to bypass agent safety filters or override core instructions.
Ingestion points: Data input for validation functions documented in SKILL.md and references/api-reference.md.
Boundary markers: The skill focuses on data-level validation but does not provide specific markers for LLM context isolation.
Capability inventory: Data validation, coercion, and transformation within a Clojure environment.
Sanitization: The library documented is specifically designed for data sanitization and schema enforcement.

clojure-malli