meta-decompile

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted XML data through the ObjectPath parameter, which could be exploited to inject instructions into the agent's context when it processes the resulting JSON output.
  • Ingestion points: XML file specified by ObjectPath in scripts/meta-decompile.ps1 and scripts/meta-decompile.py.
  • Boundary markers: The skill does not provide clear boundary markers or instructions to the agent to ignore embedded commands in the processed data.
  • Capability inventory: The skill has the ability to read files and return their contents to the agent context using Bash and Read tools.
  • Sanitization: While the output is valid JSON, no content sanitization is performed to prevent downstream prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 10:29 PM
Security Audit — agent-trust-hub — meta-decompile