A plugin and command-line tool for LLM instruction files that detects overly broad semantic wording and provides protected Hooks, project-level local rule injection, and semantic trap checks for Skills, Prompts, and Agents.
2 files for Codex, OpenCode and Claude Code: semantic-linter AGENTS.md, semantic-linter CLAUDE.md — 663 tokens loaded in every session.
AGENTS.md A 99 tok CLAUDE.md A 564 tok These files are SummerSec/semantic-linter's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.