Open-source enforcement at runtime that masks sensitive data and blocks destructive commands from AI agents. You change one thing: your agent's port.
Latest release 1.272.0 · 26 Oct 2026
11 files for Claude Code: PostToolUse, hoop, hoop CLAUDE.md, qa and 7 more — 8,033 tokens loaded in every session.
CLAUDE.md A 8,033 tok .claude/settings.json A — .claude/settings.json A — .claude/settings.json A — .claude/settings.json A — .claude/skills/build-ticket/SKILL.md A 58 tok .claude/skills/evidence/SKILL.md A 75 tok .claude/skills/shape/SKILL.md A 47 tok .claude/skills/gcp-marketplace-publish/SKILL.md A 83 tok .claude/skills/support-agent/SKILL.md A 62 tok .claude/agents/qa.md A 33 tok These files are hoophq/hoop's own configuration — they tell 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.