apple-crash-symbolication

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted .ips crash logs provided by the user, which creates a surface for indirect prompt injection attacks through fields like Application Specific Information (asi).\n
  • Ingestion points: User-provided .ips JSON files are parsed by the instructions in SKILL.md and the PowerShell script scripts/Symbolicate-Crash.ps1.\n
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat external crash data as potentially untrustworthy or to ignore embedded commands.\n
  • Capability inventory: The skill executes local shell commands (atos, dwarfdump, idevicecrashreport) and performs network downloads to fetch symbol files.\n
  • Sanitization: Content from the crash log fields is interpreted by the agent for analysis without escaping or validation to prevent the data from overriding agent instructions.\n- [EXTERNAL_DOWNLOADS]: Fetches symbol files from msdl.microsoft.com. This is a well-known service provided by Microsoft for legitimate development and diagnostic purposes.\n- [COMMAND_EXECUTION]: Invokes local system tools including atos, dwarfdump, and idevicecrashreport. These are standard diagnostic utilities used in Apple platform development and are invoked here for their intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 02:18 AM
Security Audit — agent-trust-hub — apple-crash-symbolication