The Agent Skills Directory

[SAFE]: The skill operates entirely within the local environment to transform JSON data into document files. It does not perform any network operations or access sensitive system configuration files. No malicious patterns or security risks were identified.
[COMMAND_EXECUTION]: The skill uses Python scripts (scripts/pptx_builder.py, scripts/json_validator.py, and scripts/pptx_validator.py) to perform its core functions. These scripts are invoked via command-line arguments and interact with the file system only to read input data and write the resulting presentation files. The scripts utilize standard library functions for JSON parsing and file handling, with no use of unsanitized subprocess calls or arbitrary command execution.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external JSON content used to populate PowerPoint slides, presenting a data ingestion surface.
Ingestion points: External data enters the context through the --input file path in the scripts/json_validator.py and scripts/pptx_builder.py scripts.
Boundary markers: The scripts/json_validator.py script enforces a specific schema and structure on the incoming data, effectively limiting the data to a predefined format.
Capability inventory: The skill's capabilities include local file system reads and writes, specifically for generating .pptx documents as implemented in scripts/pptx_builder.py.
Sanitization: Validation logic in the scripts checks for data types and required fields, ensuring the content is processed as structured data for document generation rather than as executable logic.

pptx-generator