code-standards

Warn

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The SKILL.md file contains commands using bun x ultracite. This executes a package from a public registry that is not included in the trusted organizations or repositories list. This represents an unverified dependency risk as the package is downloaded and executed at runtime.
  • [COMMAND_EXECUTION] (MEDIUM): The skill invokes system commands (bun x ultracite fix, check, and doctor) to perform its primary tasks. These commands execute code from the downloaded package on the host system.
  • [PROMPT_INJECTION] (LOW): The skill acts as a code reviewer, which is a surface for indirect prompt injection. Malicious instructions hidden in user-provided code (the untrusted data) could attempt to influence the agent's behavior during analysis.
  • Ingestion points: Code files reviewed for standards (referenced in SKILL.md).
  • Boundary markers: Absent; no delimiters or ignore-instructions markers found in the rule files.
  • Capability inventory: Subprocess execution via bun x (SKILL.md).
  • Sanitization: None detected in the provided rule set.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 18, 2026, 11:46 PM