code-polish

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code files from a Git repository, creating an indirect prompt injection surface. * Ingestion points: Local source files identified through Git diffs or user-supplied paths in SKILL.md. * Boundary markers: The skill instructs the agent to resolve a fixed scope and use language-specific review profiles, providing a structured execution context. * Capability inventory: The agent is authorized to modify files and execute commands such as go test, cargo test, and npm as listed in language profiles. * Sanitization: No content filtering is performed as the skill is designed for raw code analysis.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to verify code changes. * Evidence: The go.md, rust.md, and typescript.md profiles specify commands like go test, cargo test, and npm to prove behavior parity and correctness.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:30 AM
Security Audit — agent-trust-hub — code-polish