The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard Android development tools, specifically the Android Debug Bridge (ADB) and the Gradle build system. These commands are used for legitimate diagnostics such as reading logs (adb logcat), pulling traces (adb pull), and inspecting build dependencies (./gradlew dependencies).
[DATA_EXPOSURE]: The instructions involve accessing application-specific data, including internal storage via adb shell run-as, logcat output, and heap dumps. While this data can be sensitive, its access is limited to the application being debugged and is a standard requirement for effective root-cause analysis in mobile development.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by consuming untrusted data from external sources such as application logs, ANR traces, and UI layout trees.
Ingestion points: Data enters the agent context through adb logcat, adb pull /data/anr/traces.txt, and the android layout command.
Boundary markers: No specific boundary markers or "ignore instructions" warnings are defined for the processed data.
Capability inventory: The skill possesses the capability to execute shell commands (adb, gradlew) and file system operations.
Sanitization: There are no explicit instructions for sanitizing or escaping the content of logs or traces before analysis.

android-debugging