cad-dxf-agent
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the
cad-ai-agentPython package from the author's GitHub repository (github.com/jeremylongshore/cad-ai-agent.git) usingpipif the utility is not already installed on the system. - [COMMAND_EXECUTION]: The skill uses the
Bashtool to executecad-analyzeandcad-revisioncommands. These tools are used to perform deterministic analysis of DXF drawings, such as compliance checks, health audits, and quantity takeoffs. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external DXF files.
- Ingestion points: DXF files located via
Globor provided by the user. - Boundary markers: None identified in the prompt instructions.
- Capability inventory:
Bash(command execution),Read, andGlobtools are accessible. - Sanitization: Findings are parsed from JSON output and reported in prose, which reduces but does not entirely eliminate the risk of the agent interpreting embedded malicious instructions.
Audit Metadata