The Agent Skills Directory

[SAFE]: The skill serves a legitimate engineering purpose and does not exhibit any malicious patterns or security risks across the evaluated categories.
[DATA_EXPOSURE]: No network calls, credential handling, or sensitive file access operations were found. The Python script only uses standard input/output for calculations.
[REMOTE_CODE_EXECUTION]: The skill does not download external code or scripts. The provided Python script selector.py is local and relies solely on standard library imports (math, typing, dataclasses, sys).
[PROMPT_INJECTION]: Documentation in SKILL.md and reference.md contains purely informational content with no attempts to override agent behavior or bypass safety guidelines.
[COMMAND_EXECUTION]: No shell commands or subprocess spawning are present. User input is strictly cast to float types, preventing command injection through the CLI tool.

pump-selection-helper