Stop That Shit is a set of hooks and skills that guard AI coding agents against unrequested work such as unnecessary checks, files, dependencies, or task-scope expansion. It is designed for Codex, Claude Code, OpenCode, Hermes Agent CLI, and Pi workflows, and its catalogue entries are the project's own guards, skills, and plugin.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add lennney/stop-that-shit --skill stssgit clone --depth 1 https://github.com/lennney/stop-that-shitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lennney/stop-that-shit/stss)<a href="https://agentmods.dev/skills/lennney/stop-that-shit/stss"><img src="https://agentmods.dev/badge/skills/lennney/stop-that-shit/stss.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00063 | $0.00766 |
| Opus 5 | $0.00032 | $0.00383 |
| Sonnet 5 | $0.00013 | $0.00153 |
| Haiku 4.5 | $0.00006 | $0.00077 |
Grade A, and why
stss scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stop That Shit Slop
别再废话。
Make the point without writing for an imaginary critic. Preserve any limitation that changes the reader's decision.
Stay in scope
Reduce defensive wording in decision-facing text. Do not classify text as AI-written, promise to remove an "AI voice," humanize unrelated prose, or perform a general style cleanup.
Judge sentences by their decision use, not by who or what may have written them.
Select the mode
rewrite: Return a tighter version. This is the default when the user provides text without a mode.audit: Identify defensive writing and propose the smallest useful fix. Do not rewrite the full text.
Follow an explicit mode. Infer the mode from the request only when none is given.
Record the claims
Before editing, build a private Claim Ledger from the supplied material:
- facts and numbers;
- named actors and sources;
- evidence strength;
- stated uncertainty and scope;
- the action requested from the reader.
Do not print the ledger unless the user asks for it. Do not add facts, sources, numbers, certainty, or causality.
Test each defensive sentence
Apply the Sentence Consumer Test:
- Who will use this sentence?
- What decision does it change?
- What becomes false or misleading if it is removed?
Choose one action:
DROP: It only narrates diligence, anticipates criticism, apologizes, or says what the document does not attempt.CALIBRATE: It contains real uncertainty but uses stacked hedges. Keep one precise limitation.RELOCATE: It matters, but interrupts the main claim. Move it next to the affected claim or decision.KEEP: It changes a legal, safety, financial, methodological, contractual, or scope decision.
Do not decide from a keyword alone. The same phrase can be waste in one context and necessary in another.
Rewrite
- Lead with the requested claim, decision, or proposal.
- State scope positively: say what the work covers.
- Replace stacked hedges with one evidence-matched qualifier.
- Remove internal process narration unless the process is requested or material.
- Put a necessary limitation beside the claim it limits.
- Compare the result with the Claim Ledger.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 90 lines · 63 tokens per session scan A 45572e089d43
stss is a skill published in the GitHub repository lennney/stop-that-shit (1,629 stars, last pushed 3d ago), licensed MIT. It adds 63 tokens to every session and 766 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
grok-build-delegate
Use the local Grok Build CLI for second-opinion reviews, adversarial critiques, or delegated read-only investigation from Codex.
multitask-run
Split a task into parallel Codex workers and execute it.
multitask-cancel
Cancel a running Codex multitask job.
multitask-plan
Produce a Codex multitask worker plan without executing workers.
multitask-result
Show the final report for a Codex multitask job.
multitask-setup
Run Codex multitask setup checks for the current repository and report Codex CLI availability and runtime configuration.