The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze untrusted data (APK/IPA files, decompiled source code) which provides a surface for indirect prompt injection. A malicious application could contain hidden instructions designed to trick the agent into performing unauthorized actions during the analysis process.
Ingestion points: Decompiled Android source code (JADX/Apktool output) and iOS application binaries.
Boundary markers: No specific delimiters or warnings to ignore embedded instructions are provided.
Capability inventory: The skill uses powerful tools including grep, frida, docker, and semgrep to process and interact with the application data.
Sanitization: No sanitization or validation mechanisms for the analyzed application content are defined.
[EXTERNAL_DOWNLOADS]: The skill provides commands to download and run the Mobile Security Framework (MobSF) using a Docker image (opensecurity/mobile-security-framework-mobsf). MobSF is a well-known open-source security tool, and this download is documented as part of the intended security auditing workflow.
[COMMAND_EXECUTION]: The skill makes extensive use of CLI tools for mobile analysis, including jadx, apktool, frida, and semgrep, which are standard for the described security assessment tasks.

mobile-weak-crypto