skills/zeij-drive/deobf/deobf-all/Gen Agent Trust Hub

deobf-all

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill's primary function is to process untrusted external data—such as native binaries, JavaScript malware, and Android APKs—which establishes a significant surface for indirect prompt injection attacks.
  • Ingestion points: The skill explicitly instructs the agent to read and triage external target files using tools like read_file, apktool, and jadx (referenced in SKILL.md).
  • Boundary markers: The dispatcher does not provide instructions for using delimiters or protective wrappers to prevent the agent from accidentally executing instructions found within the analyzed files.
  • Capability inventory: The dispatcher enables access to powerful sub-skills that perform dynamic instrumentation (frida-17), symbolic execution (symbolic-execution-tools), and automated script execution.
  • Sanitization: There are no provisions for sanitizing, escaping, or validating the content of the target files before they are loaded into the agent's context for analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 02:56 AM
Security Audit — agent-trust-hub — deobf-all