The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a local Python script scripts/quality_gate.py and instructs the agent to run it to verify output quality. This script uses standard libraries (re, pathlib) to perform a regex-based check against a list of banned terms and stylistic requirements to ensure the output adheres to the defined persona.
[DYNAMIC_EXECUTION]: The test suite scripts/test_quality_gate.py utilizes importlib.util to dynamically load the quality gate script for unit testing purposes. This is a standard development practice for verifying the utility's logic.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user data in the form of product management problems. It mitigates potential injection risks by using a structured reasoning framework and an automated quality gate script that validates the agent's response against defined safety and stylistic constraints before presenting it to the user.
[EXTERNAL_DOWNLOADS]: The documentation references several external GitHub repositories as methodology sources. Analysis confirms these are for informational purposes only; the skill does not contain instructions to perform automated downloads or execute remote code from these locations.

product-decision-agent