The Agent Skills Directory

[SAFE]: The skill serves as a protocol definition file for an agent framework. It provides instructions on operational modes (Build, Incident, Experiment), cognitive integrity (using specific MCP tools like sequentialthinking and context7), and technical standards (TDD, surgical code changes). No evidence of prompt injection, data exfiltration, or unauthorized command execution was found. The security section explicitly mandates treating external inputs as hostile and redacting sensitive information.

product-manager