The Agent Skills Directory

[SAFE]: The skill functions as a technical resource for TYPO3 developers, providing best practices for CKEditor 5 migration and plugin architecture. It includes a local shell script for verifying extension structure which does not perform any network operations or access sensitive system files.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it analyzes project source code and YAML configuration files via automated checkpoints. While this allows external data to enter the agent's context, the risk is minimal due to the skill's rigid, non-interpolated execution patterns.
Ingestion points: The agent reads content from project files located in */JavaScript/[Cc]keditor* and */Configuration/RTE/*.yaml during analysis.
Boundary markers: The checkpoints.yaml file includes explicit instructions and constraints to prevent command injection, warning against the use of command substitution, backticks, and shell operators like || or &&.
Capability inventory: The skill utilizes the agent's ability to execute shell commands such as find, grep, xargs, and head to perform structural validation.
Sanitization: Command patterns are statically defined in the checkpoints, avoiding the interpolation of file content directly into shell strings, thereby preventing typical command injection vectors.

typo3-ckeditor5