The Agent Skills Directory

[SAFE]: The skill provides legitimate manufacturing algorithms for calculating process capability indices. The implementation in scripts/cpk.py uses standard mathematical libraries and does not perform any dangerous operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided measurement data which constitutes a theoretical data ingestion surface, though no exploitable vulnerabilities were found.
Ingestion points: The scripts/cpk.py utility accepts process measurement data through a JSON file via the --input flag.
Boundary markers: The script expects structured data in JSON format and utilizes the standard argparse library for command-line validation.
Capability inventory: The skill's capabilities are restricted to mathematical processing and returning JSON results. It has no access to network tools, file-writing capabilities, or shell execution tools.
Sanitization: Inputs are parsed using the standard json library and cast to numerical types (floats), effectively neutralizing any embedded command strings or non-numeric instructions.

algo-mfg-cpk