The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external Excel files, creating a surface for indirect prompt injection attacks.
Ingestion points: The scripts scripts/read_xlsx.py and scripts/xlsx_to_json.py use the openpyxl library to read content from user-specified or environment-provided .xlsx files.
Boundary markers: The skill instructions do not specify any boundary markers or delimiters (e.g., XML tags or clear 'ignore embedded instructions' warnings) when presenting the extracted spreadsheet data to the agent.
Capability inventory: The agent executes shell commands (wsl python3) to run the extraction scripts. The resulting text is then processed by the agent, which may have the capability to write files, execute further commands, or access network resources depending on its global configuration.
Sanitization: The scripts perform basic string conversion and Markdown formatting, but they do not include logic to detect or sanitize malicious prompt injection attempts embedded within cell values.
[EXTERNAL_DOWNLOADS]: The skill documentation instructs the user/agent to install the openpyxl package from a public registry (PyPI). While openpyxl is a well-known and standard library for Excel processing, users should ensure it is installed in a controlled environment.

xlsx-reader