glb-web-handoff

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local tools blendops glb-inspect and blendops glb-handoff which map to scripts/inspect-glb.mjs and scripts/generate-glb-handoff-report.mjs for asset validation.
  • [DYNAMIC_EXECUTION]: The skill invokes Node.js scripts via static relative paths to perform automated GLB container inspection and manifest generation as part of its operating procedure.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted GLB binary files and metadata from upstream validation skills. (1) Ingestion points: GLB 2.0 files and structured manifest JSON data from render-export-evidence and scene-readiness. (2) Boundary markers: The skill employs truth-state laws and downgrade-only logic to manage claims, but does not define explicit data-level delimiters. (3) Capability inventory: Includes execution of local Node.js scripts and delegation of tasks to other specialized skills. (4) Sanitization: No explicit content sanitization or validation routines for input data are described.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 03:04 PM