The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted external data by ingesting user-provided content to generate infographic prompts.
Ingestion points: Source content is read from a user-specified file path or pasted text into source.md (Step 1.2).
Boundary markers: The base-prompt.md template lacks clear delimiters or instructions to ignore embedded commands within the {{CONTENT}} variable.
Capability inventory: The skill can execute shell commands (test, mv), write multiple files to the local system, and invoke other agent skills (image generation).
Sanitization: No input validation or escaping is applied to the source content before it is used in the prompt generation process.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands to manage its environment and configuration.
The agent is instructed to use test -f to check for the existence of EXTEND.md in both the project root and the user's home directory within a vendor-specific hidden folder (.tuzi-skills).
Automated backup procedures involve renaming existing files, implying the use of filesystem manipulation commands like mv.

tuzi-infographic