td-working-mode
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected. The skill provides architectural posture and best practices for TouchDesigner development within the agent's toolset.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external project data using tools like
get_dat_contentandget_parameters, then write code to the local filesystem viaset_dat_content. - Ingestion points:
get_dat_content,inspect_values, andget_parameters(SKILL.md). - Boundary markers: None mentioned.
- Capability inventory:
set_dat_contentallows writing files to the local project directory. - Sanitization: None mentioned. While this represents a standard data ingestion surface for a development tool, the skill focuses on legitimate workflow guidance and does not contain exploitable patterns or malicious instructions.
Audit Metadata