The Agent Skills Directory

[SAFE]: The skill is primarily composed of Markdown documentation, playbooks, and templates. An exhaustive review found no evidence of prompt injection, data exposure, or obfuscation techniques. All logic is grounded in established product management literature.
[COMMAND_EXECUTION]: A shell script (evals/run-static-checks.sh) is provided for automated linting and validation of the skill's internal structure. This script performs benign local operations such as verifying file presence, checking word counts, and parsing its own internal CSV router using a localized Python snippet. It does not engage in network communication or access sensitive user environment data.

product-discovery