skills/vllnt/skills/improve-tests/Gen Agent Trust Hub

improve-tests

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to identify and execute native test commands and coverage instrumentation specific to the repository being analyzed. This capability is central to the skill's purpose of measuring test performance and verifying that code changes preserve existing protection.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes test source files, assertions, and repository configurations which could potentially contain untrusted instructions if the repository itself is malicious. * Ingestion points: Reads test source code, repository metadata, and discovery configurations from the local file system. * Boundary markers: The instructions do not define specific delimiters to isolate ingested code content from the agent's internal reasoning. * Capability inventory: The skill utilizes shell execution for test runners and file system writes for refactoring tasks. * Sanitization: The workflow incorporates a structured quality validation protocol that includes self-assessment and verification of detection efficacy (e.g., negative controls) to ensure the integrity of the refactoring process.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 08:18 PM