ifc-core-ifcx-architecture

Pass

Audited by Gen Agent Trust Hub on Aug 9, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection attack surface.
  • Ingestion points: The skill instructs the agent on how to parse and interpret .ifcx files in SKILL.md ('How to approach an .ifcx file') and references/methods.md ('2. The .ifcx File: Structural Summary').
  • Boundary markers: The instructions do not define clear delimiters or directives to isolate the data contained within .ifcx files from the agent's prompt context.
  • Capability inventory: The host agent (Claude Code) possesses capabilities for shell execution and file modification that could be abused if malicious instructions are embedded in the parsed JSON data.
  • Sanitization: No sanitization, validation, or filtering logic is provided for the content of the data objects during the scene composition process.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 9, 2026, 04:40 PM
Security Audit — agent-trust-hub — ifc-core-ifcx-architecture