The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple local Python scripts (compose_pptx.py, chroma_key.py, and slice_grid.py) to manage the synthesis of PPTX files and the processing of image layers. These scripts are core components used for the skill's orchestration logic.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied themes and content which are then used to generate prompts for the imagegen tool and populate data in JSON files such as outline.json and layout.json. This workflow creates an indirect prompt injection surface where untrusted input could potentially influence agent behavior or script execution.
Ingestion points: User-provided themes and content defined in SKILL.md, and structured data processed from outline.json and layout.json mentioned in references/pipeline.md.
Boundary markers: The provided documentation does not specify the use of delimiters or instructions to ignore embedded commands within the processed data.
Capability inventory: The skill uses the imagegen tool for image creation, performs file system operations (copying and writing in slides/, out/, and editable/ directories), and executes external Python scripts.
Sanitization: No explicit sanitization or validation of user-controlled content is described before the data is interpolated into prompts or processed by the local scripts.

GordenSuperPPTSkill