The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script scripts/escalate.py to automate legal fee calculations and document generation for contents-certified mail and court filings.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by ingesting data associated with invoice IDs. 1. Ingestion points: The scripts/escalate.py script likely reads external data associated with the provided [ID] to populate legal forms. 2. Boundary markers: No delimiters or explicit instructions to ignore embedded commands are documented for the data ingestion process. 3. Capability inventory: The skill possesses the ability to execute subprocesses (Python) and write files to the local .fukugyo/ directory. 4. Sanitization: There is no documentation indicating that external data is validated or sanitized before interpolation into prompts.
[SAFE]: The skill provides links to the Japanese Ministry of Health, Labour and Welfare (mhlw.go.jp) and official legal support services which are recognized as trusted sources.

fukugyo-escalate