The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a wide array of shell commands to perform security analysis. These include curl for web scraping and APK acquisition, grep for secret extraction, unzip/7z for file extraction, jadx and apktool for decompilation, openssl for certificate analysis, and adb/frida/objection for runtime instrumentation.
[EXTERNAL_DOWNLOADS]: The pipeline instructions include downloading APK files from mirrors such as APKPure (apkpure.net) and APKMirror (apkmirror.com), as well as fetching the jadx decompiler from GitHub. While these are functional requirements for the skill's purpose, they involve fetching content from external sources.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection because it ingests and processes untrusted data from external sources, specifically Play Store HTML and decompiled APK files (DEX, resources, manifests).
Ingestion points: Content is fetched via curl from the Play Store and APK mirrors, then extracted and decompiled using unzip, 7z, and jadx (SKILL.md).
Boundary markers: The skill includes a 'Mandatory confirmation gate' requiring the user to confirm target scope and commands, providing a human-in-the-loop check.
Capability inventory: The skill uses curl, grep, unzip, jadx, adb, frida, and objection (SKILL.md).
Sanitization: No explicit sanitization or filtering is described for the text extracted from APK files before the agent processes it.

apk-redteam-pipeline