The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses openpyxl via a Python code template to generate Excel spreadsheets. While the provided code is a template, the agent will execute this logic using its internal Python environment to create the .xlsx file. This is a standard and intended function for report generation.
[DATA_EXPOSURE]: The skill instructions specify reading from /mnt/user-data/uploads/ (for user-uploaded files) and writing to /mnt/user-data/outputs/. This is standard behavior for file-processing skills and does not involve accessing sensitive system credentials or private keys.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and summarize external data from URLs, PDFs, or pasted text. This creates an attack surface where malicious instructions hidden in the analyzed documents could attempt to influence the agent's behavior. The skill lacks specific boundary markers or 'ignore' instructions for the processed content.
Ingestion points: Reads content from URLs, user-pasted text, and files in /mnt/user-data/uploads/ (specified in SKILL.md).
Boundary markers: None present. The instructions do not tell the model to ignore embedded instructions in the source text.
Capability inventory: The skill has the capability to generate and write .xlsx files using openpyxl in the Python tool environment.
Sanitization: The skill instructions include a quality check to ensure columns are populated and formats are followed, but there is no explicit sanitization of the input text to prevent injection.

enforcement-action-analysis-amir-fadavi