charted-scaffold

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill processes external design documents provided via $ARGUMENTS[0] without explicit boundary markers or sanitization.
  • Ingestion points: The content of the file at designDocPath is interpolated into the task prompt.
  • Boundary markers: Absent; the document content is not delimited from the system instructions.
  • Capability inventory: The skill is capable of generating code that may be integrated into a codebase. However, it is instructed to make all new logic throw "work in progress" errors, which acts as a safety control against executing malicious logic hidden in a design doc.
  • Sanitization: None provided for the ingested document text.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 12:54 PM