cost-booster-edit

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a Bash-invoked Node.js command to execute its transformation logic. It passes code and intent data as command-line arguments, which is a standard method for interacting with CLI tools.
  • [PROMPT_INJECTION]: The skill processes file content and transformation intents, which constitutes a surface for indirect prompt injection. This risk is inherent to code-transformation tasks and is mitigated by a confidence-scoring system that rejects transformations with a confidence level below 0.5.
  • Ingestion points: Reads source code from local files and accepts transformation 'intents' as arguments.
  • Boundary markers: None identified in the prompt templates.
  • Capability inventory: File system access (read/write) and shell execution via Bash.
  • Sanitization: Uses the AgentBooster engine's confidence scoring to validate transformations before writing back to the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 01:23 PM
Security Audit — agent-trust-hub — cost-booster-edit