explore-codebase

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Command Execution] (SAFE): The skill is configured to use industry-standard tools (rg, fd, sg) via the Bash tool specifically for codebase exploration. All provided examples and documentation align with legitimate developer workflows for finding and tracing code.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded secrets, credentials, or unauthorized network operations were found. The tool functions strictly on local data provided to the agent for analysis.
  • [Indirect Prompt Injection] (SAFE): While the skill is designed to ingest and process untrusted codebase data (Category 8 surface), this is its primary intended function. The instructions do not facilitate the execution of instructions found within that data, and the capability tier is limited to search operations.
  • [Obfuscation] (SAFE): No encoded content, zero-width characters, or homoglyph-based evasion techniques were identified in any of the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 01:10 PM