skills/impertio-studio/thatopen-claude-skill-package/thatopen-agents-model-analyzer/Gen Agent Trust Hub
thatopen-agents-model-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes IFC models, which are external data sources, creating a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through FragmentsManager.getData and ifcApi query methods as seen in SKILL.md, references/methods.md, and references/examples.md.
- Boundary markers: No explicit delimiters or boundary warnings are specified in the report templates in SKILL.md.
- Capability inventory: Capabilities demonstrated in SKILL.md and references/examples.md are limited to data extraction and report generation; no shell execution, file system modifications, or network exfiltration operations are present.
- Sanitization: The skill does not provide mechanisms for sanitizing or validating text strings extracted from the IFC models before report generation.
Audit Metadata