plankton-code-quality

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill describes the use of PreToolUse and PostToolUse hooks to execute local shell scripts (multi_linter.sh, protect_linter_configs.sh) and various CLI tools including ruff, biome, shfmt, and hadolint upon every file modification.
  • [COMMAND_EXECUTION]: It automates the spawning of claude subprocesses (utilizing Haiku, Sonnet, or Opus models) to perform autonomous code repairs based on linter output.
  • [COMMAND_EXECUTION]: The skill enforces a specific development environment by using hooks to block the execution of standard package managers like npm, yarn, and pip, forcing the use of bun or uv instead.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface; because it passes code fragments containing linter violations to a sub-agent subprocess, malicious instructions embedded within code comments or strings in the files being edited could theoretically influence the behavior of the sub-agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:47 PM
Security Audit — agent-trust-hub — plankton-code-quality