byterover-explore

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill implements best practices for security by providing explicit instructions to skip reading sensitive files such as .env, .key, credentials.json, and other similar files during the exploration process.
  • [COMMAND_EXECUTION]: The skill makes use of the brv (ByteRover) CLI tool to perform its operations, including brv status, brv query, and brv curate. These commands are used as intended for codebase analysis and knowledge management.
  • [PROMPT_INJECTION]: The skill possesses an inherent attack surface for indirect prompt injection because it reads and processes untrusted data from the codebase being analyzed.
  • Ingestion points: Project files like package.json, README.md, and source code files accessed through the brv curate -f flag.
  • Boundary markers: None explicitly defined in the skill instructions to separate processed data from instructions.
  • Capability inventory: The skill can execute CLI commands and read file system contents via the ByteRover tool.
  • Sanitization: There are no explicit sanitization or content-escaping steps described for the data being ingested from the codebase.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 03:21 PM
Security Audit — agent-trust-hub — byterover-explore