The Agent Skills Directory

[SAFE]: The skill instructions define a standard engineering process for implementation and refactoring. No malicious patterns, hidden commands, or security vulnerabilities were detected in the instructions or referenced documents.
[COMMAND_EXECUTION]: The skill includes instructions to run verification commands and build local engineering tools, such as scripts or harnesses, as part of the intended implementation process. These operations are aligned with standard development practices and are restricted to the repository scope.
[PROMPT_INJECTION]: No behavior override markers or attempts to bypass safety filters were found. The skill emphasizes engineering discipline and includes positive security heuristics, such as parsing untrusted input at boundaries.

implement-work