#134134th of 259 in Skills, plugins and prompts
stop-that-shit
Skill and guard that stop coding agents from scope creep and needless defensive code
- Stars
- 2.9k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- JavaScript
79 on 11 Aug 20262,942 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Stop That Shit is a skill plus a tool-call guard for coding agents (Codex, Claude Code, OpenCode, Hermes Agent CLI, Pi, Oh My Pi). The skill sets rules against scope creep, unused hashing, read-only violations and repeated work. The guard blocks writes in review/answer/monitor modes, writes outside files=, new dependencies, extra subagents and recognizable hash operations unless allowed with flags like deps=allow or hash=allow.
Who it is for: Developers who run coding agents and want to enforce task boundaries
Strengths
- Per-task contract with modes (review, answer, monitor, change, watch) and files=, deps=, hash=, agents= flags
- Covers six agent hosts, each with install steps, plus a skill-only option
- Guard checks are narrow and documented: writes, dependencies, subagents, hashes, paths
- Ships 18 public Bad/Good cases and an evidence file with paired runs
Weaknesses
- Guard stays inactive until a mode is used; default state is OBSERVING / unconfirmed
- A deny response does not prove the host skipped the action; isolation is left to the host sandbox
- Hash detection covers only recognizable operations; other over-engineering relies on skill judgment
- Main README is Chinese; English and Korean versions are separate files
What it needs
- no GPU
- Needs Node.js 18+, Node.js 22.19+ (Pi 0.84.4)
- Models: Codex, Claude Code, OpenCode 1.18.18+, Hermes Agent CLI, Pi Coding Agent 0.84.4
Also in Skills, plugins and prompts
See all 259| Rank | Project | Score |
|---|---|---|
| 1 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 96 out of 100 |
| 2 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 95 out of 100 |
| 3 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| #4 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 94 out of 100 |
| #5 | oh-my-claudecodeMulti-agent orchestration plugin and CLI for Claude Code | 94 out of 100 |