enhance-cross-file

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface.
  • Ingestion points: Processes workspace files including agents, skills, and workflows as mentioned in the analysis purpose.
  • Boundary markers: Absent; no specific delimiters or instructions are provided to the agent to disregard instructions within the analyzed data.
  • Capability inventory: Execution of a Node.js script to process files.
  • Sanitization: No sanitization of input data is documented.
  • [COMMAND_EXECUTION]: The skill executes a local file at ./lib/enhance/cross-file-analyzer.js using the node -e command. The source code for this script is not provided in the skill package, preventing a full security audit of the performed operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:48 PM