Related Servers

Most tools target a distinct resource plus action (class source vs. smali vs. methods vs. fields vs. xrefs), and the rename_* family is cleanly scoped per symbol. A few pairs are close: fetch_current_class vs. get_class_source, and get_method_by_name vs. search_method_by_name, but the descriptions (fetch one method by class/name vs. search names across all classes) resolve them adequately. get_main_application_classes_names vs. get_all_classes is also slightly fuzzy but the manifest-package scope distinguishes it.

Predominantly consistent snake_case verb_noun (get_class_source, get_fields_of_class, search_classes_by_keyword, rename_method). Minor deviations: a fetch_ vs. get_ mix, and the debug tools use a debug_ prefix group rather than the verb_noun form used elsewhere. Still highly predictable overall.

32 tools is on the heavy side and the surface spans four sub-domains (code retrieval, search, resources/manifest, rename refactoring, plus a debugger). There is some redundancy, e.g. get_android_manifest, get_manifest_component, get_main_activity_class, and get_main_application_classes_names/code all read from the same manifest and could be consolidated. Each area is individually justified, but the set feels broader than a single coherent scope.

Coverage is strong for reverse-engineering workflows: class/method/field retrieval, smali, three kinds of xrefs, keyword search, resources, manifest, rename refactoring, and cache management. Gaps exist in the debugger subset, which only exposes read-only inspection (stack frames, threads, variables) with no continue/step/breakpoint control, and there is no smali-for-method or export/dump operation. These are workable gaps rather than dead ends.