The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes common shell commands such as mkdir, rsync, cp, and python3 to set up its directory structure and execute local processing scripts. These operations are restricted to the skill's own workspace and local files provided by the user.
[INDIRECT_PROMPT_INJECTION]: The skill processes external untrusted data including billing exports (CSV, FOCUS) and invoices (PDF, email).
Ingestion points: External data is ingested through various formats mentioned in the Inputs section of SKILL.md, such as invoices, billing exports, and cloud cost exports.
Boundary markers: No specific boundary markers or instructions to ignore embedded content within the data files are defined in the skill instructions.
Capability inventory: The skill possesses capabilities for file system management (mkdir, cp, rsync), execution of local Python scripts (python3), and the ability to write documentation pages.
Sanitization: Sanitization and validation of the external content are performed by the scripts/focus_export.py script, which checks and converts the data into a standardized format before baseline generation.

cost-baseline