The Agent Skills Directory

[SAFE]: The skill consists of documentation and code examples that promote secure development practices within the CakePHP framework. It explicitly identifies and warns against insecure anti-patterns such as SQL string interpolation and mass assignment, providing safe, parameter-bound alternatives.
[PROMPT_INJECTION]: The skill defines patterns for handling untrusted user input within database operations and documents necessary framework-level protections.
Ingestion points: Untrusted user input is identified in database query conditions and request data used to populate entities (SKILL.md, references/examples.md).
Boundary markers: The instructions explicitly require the use of array-based conditions to enforce the ORM's automatic parameter binding (SKILL.md, references/recipes.md).
Capability inventory: The skill allows for performing database reads, writes, and aggregations using Table and Entity objects (SKILL.md).
Sanitization: The documentation mandates the use of the framework's built-in parameter binding and the $_accessible entity property to prevent SQL injection and unauthorized field updates (references/core.md, references/examples.md).

orm