The Agent Skills Directory

[DYNAMIC_EXECUTION]: Both scripts/batch_generate.py and scripts/infographic_cli.py use importlib.util to dynamically load and execute a generator module from a computed relative path (../../visual-design-system/scripts/generate_infographic.py). Loading and executing code from paths determined at runtime is a risky pattern that could be exploited to execute arbitrary code if the file at that location is replaced or manipulated.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data via JSON and YAML configurations and command-line arguments to populate infographic templates. This creates a vulnerability surface where malicious instructions could be embedded in the processed data.
Ingestion points: Configuration files loaded in scripts/batch_generate.py (--config) and data strings or files in scripts/infographic_cli.py (--data, --data-file).
Boundary markers: Absent; there are no clear delimiters or instructions to ignore embedded commands in the processed data.
Capability inventory: The skill has capabilities for file system writes, directory creation, and dynamic code execution.
Sanitization: While standard parsers like json.loads and yaml.safe_load are used, they only validate structure and do not sanitize the actual content against prompt injection or malicious payloads.
[COMMAND_EXECUTION]: The SKILL.md documentation includes multiple examples of executing shell commands to run the infographic generation scripts. This facilitates command-line operations that could lead to command injection if the agent interpolates unsanitized user input into these shell commands.

infographic-generator