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 rodrigohighermind/highermind-code-skills --skill hm-llm-guardrailsgit clone --depth 1 https://github.com/rodrigohighermind/highermind-code-skillsWrote 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/rodrigohighermind/highermind-code-skills/hm-llm-guardrails)<a href="https://agentmods.dev/skills/rodrigohighermind/highermind-code-skills/hm-llm-guardrails"><img src="https://agentmods.dev/badge/skills/rodrigohighermind/highermind-code-skills/hm-llm-guardrails/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/rodrigohighermind/highermind-code-skills/hm-llm-guardrails"><img src="https://agentmods.dev/badge/skills/rodrigohighermind/highermind-code-skills/hm-llm-guardrails.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00149 | $0.03695 |
| Opus 5 | $0.00075 | $0.01847 |
| Sonnet 5 | $0.00030 | $0.00739 |
| Haiku 4.5 | $0.00015 | $0.00369 |
Grade A, and why
hm-llm-guardrails 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 11d 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 — 370 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/hm-llm-guardrails — Patterns LLM-app (v2)
Você está agora em modo LLM guardrails. Seu trabalho e validar que app que integra LLM (Claude, GPT, Gemini, Llama via API) tem todos os guardrails de produção em pe. Não esta otimizado pra cobrir um caso de uso. Cobre TODOS os patterns recorrentes.
Princípio central
LLMs sao componentes não-deterministicos com custo variavel e latencia variavel, em provedores externos. Tratar como qualquer outra dependência externa: rate limit, retry, fallback, timeout, dedupe, observabilidade. Sem guardrails, você não tem produto — tem demo.
Quando usar
- Antes de shippar feature LLM pra produção
- Quando custo de API explode sem explicacao
- Quando user reclama de "respostas demoram demais", "trava no meio", "respondeu coisa errada"
- Quando adicionar tool calling, agentes, ou chat persistente
- Apos integrar provider novo (multi-provider routing)
Patterns obrigatorios
1. Sliding window de chat history
Problema: chat history sem limite cresce sem limite. Conversa de 50+ turns estoura context window (200k tokens em Sonnet 4.6 / 1M com beta header) ou explode custo.
Pattern:
// SQL/ORM: últimas N messages, ordem cronologica preservada
const recent = await db
.select()
.from(messages)
.where(eq(messages.threadId, id))
.orderBy(desc(messages.createdAt))
.limit(CHAT_HISTORY_LIMIT) // tipicamente 30
const history = recent.reverse()
Limite típico: 30 turns (15 user/assistant pares). Suficiente pra contexto recente, longe do limite de tokens.
Avancado: Sliding window + summary das mensagens antigas (compactacao). Soma das duas no prompt = janela infinita efetiva. Custa 1 LLM call extra periodicamente pra gerar summary.
2. Lazy client factory (sem singleton stale)
Problema: SDK client (Anthropic, OpenAI) instanciado no module-load com getApiKey(). Se user trocar key em runtime (settings page), continua usando a antiga até restart.
Pattern:
let cachedClient: { key: string; client: Anthropic } | null = null
export class ApiKeyMissingError extends Error {}
export function getAnthropic(): Anthropic {
const key = getApiKey()
if (!key) throw new ApiKeyMissingError()
if (cachedClient && cachedClient.key === key) return cachedClient.client
cachedClient = { key, client: new Anthropic({ apiKey: key, timeout: 90_000, maxRetries: 1 }) }
return cachedClient.client
}
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.
- 11d ago First seen · 370 lines · 149 tokens per session scan A 02675d60cb5a
hm-llm-guardrails is a skill published in the GitHub repository rodrigohighermind/highermind-code-skills (193 stars, last pushed 2mo ago), licensed MIT. It adds 149 tokens to every session and 3,695 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
seedance-vocab-en
This skill should be used when an English Seedance 2.0 prompt needs clearer production wording, less generic prose, or precise vocabulary for camera, lighting, motion, VFX, audio, and constraints. Route blocked prompts through seedance-filter for context and boundary review.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…