The Agent Skills Directory

[PROMPT_INJECTION]: The skill documents an interface that ingests external, potentially untrusted data for report generation, which creates an indirect prompt injection attack surface.
Ingestion points: Data is entered through the ReportGenerator constructor, methods like add_table, add_text, set_data, and CLI arguments such as --csv, --data, and --config as described in SKILL.md.
Boundary markers: The documentation does not describe the use of delimiters, escaping, or explicit instructions to the model to ignore content within the data fields.
Capability inventory: The skill possesses file system write access via the save() and generate() methods, as well as file system read access for images and data source files.
Sanitization: No input sanitization, schema validation, or data filtering is mentioned in the API reference or usage instructions.

report-generator