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 NeuroverseOS/Neuroverseos-governance --skill govern-my-chatbotgit clone --depth 1 https://github.com/NeuroverseOS/Neuroverseos-governanceWrote 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/neuroverseos/neuroverseos-governance/govern-my-chatbot)<a href="https://agentmods.dev/skills/neuroverseos/neuroverseos-governance/govern-my-chatbot"><img src="https://agentmods.dev/badge/skills/neuroverseos/neuroverseos-governance/govern-my-chatbot/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/neuroverseos/neuroverseos-governance/govern-my-chatbot"><img src="https://agentmods.dev/badge/skills/neuroverseos/neuroverseos-governance/govern-my-chatbot.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00171 | $0.06468 |
| Opus 5.5 | $0.00068 | $0.02587 |
| Sonnet 5.5 | $0.00034 | $0.01294 |
| Haiku 4.5 | $0.00017 | $0.00647 |
Grade A, and why
govern-my-chatbot 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 19d 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 — 544 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Govern My Chatbot
You are helping a creator put governance on a chatbot they are building. They know what they want it to do; what they don't know is how to program it to stay within their guidelines. That gap is what this skill closes. They likely have no engineering background and may never have used Claude Code or a terminal before. They have a model — a method, a framework, a way of teaching — and they want a chatbot that teaches it faithfully, in their voice, without wandering into places it should not go.
This is a Claude Code skill. It needs to create files, install packages, and run checks in the creator's project. If you can't run commands in their project (for example, in a chat-only app), say so plainly and tell them to install Claude Code and start this skill from their project folder there.
You have two jobs at once:
- Build it right — their chatbot ends up governed by rules they own.
- Teach them — by the end, they understand what governance is, why each piece exists, and can change, test, and try to break their own rules without you.
Work through these parts in order. Do not skip ahead.
- Orientation — the problem, the idea, and this skill's own rulebook.
- Interview — draw out how the chatbot must behave, situation by situation.
- Write the rulebook — their answers become
governance/<name>.nv-world.md. - Wire it in — the app enforces the rulebook with code.
- Break it — they try to get past their own governance, and fix what they find.
- Hand over the controls — they change a rule and run every check themselves.
The two ideas to teach
1. A rule the AI can talk its way out of is a wish, not a rule. Governance means the important rules live outside the AI: in a file the creator owns, and in code that runs before and after every AI reply. The AI is handed the rules; it never gets to rewrite them. For the most serious moments (a person in crisis, say), the AI is not even consulted.
2. Rules are about meaning and behavior, not words. A rule is not a list of banned words. "Never promise results" is broken by "you'll definitely land this job", which contains no word you'd think to ban. A person in crisis may say "I don't see a way forward" rather than anything on a list. So every rule is written as what the chatbot does in a situation, and the app checks replies and messages by meaning, using a separate AI check (the checker) that reads them against the creator's rules. There are no word lists anywhere. The creator's example phrases teach the checker what a moment sounds like; they are never matched word for word.
What ships with it
10 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.
- governance/govern-my-chatbot.nv-world.md 3.8 KB
- hooks/rulebook-guard.mjs 9.4 KB runs code
- README.md 5.4 KB
- templates/break-it.ts 5.3 KB runs code
- templates/chatbot.ts 2.3 KB runs code
- templates/check-my-rules.ts 7.1 KB runs code
- templates/claude-models.ts 3.1 KB runs code
- templates/coach.nv-world.md 5.4 KB
- templates/HOW-TO-CHANGE-MY-CHATBOT.md 3.9 KB
- templates/tests.md 2.9 KB
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.
- 19d ago First seen · 544 lines · 171 tokens per session scan A 7ee63ef363ae
govern-my-chatbot is a skill published in the GitHub repository NeuroverseOS/Neuroverseos-governance (2 stars, last pushed 5d ago), licensed Apache-2.0. It adds 171 tokens to every session and 6,468 once invoked, about $0.0007 per session on Opus 5.5. 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 2026-09-29.
Other skills, from other repositories
interview-prep
Technical interview preparation expert for algorithms, system design, and behavioral questions.
goga-discover
Interview the user relentlessly about a decision until every branch of the design tree is resolved, then record the result as a short ADR.
search-lessons
Search promoted lessons for corrective actions, lifecycle state, linked rules, and linked gates.
interpret-results
Interprets AIR Blackbox scan results and maps findings to specific EU AI Act articles, recitals, and remediation steps. Use when the user has scan output and wants to understand what to fix, why it matters, or how to prioritize.
compliance-scan
Scans a Python AI project for EU AI Act compliance gaps using AIR Blackbox. Use when the user asks to check compliance, scan their code, audit their AI project, or mentions EU AI Act, Articles 9-15, or compliance checking.
smartstudy
Turn any lecture material (text, PDF, web page) into an adaptive study session — extract topics, quiz the user, track mastery with FSRS spaced repetition, and let the RL policy decide what to study next. Use when the user wants to study, revise, prepare for an exam, or asks "quiz me" on some material.