DeskcommCRM is an open-source, self-hosted customer relationship management system for businesses that sell through chat, with AI agents that handle, qualify, and sell through WhatsApp. Businesses use it to manage conversations and sales while keeping the CRM and its data on their own server. The catalogue entries provide agents, instructions, skills, and commands for working with the project.
Borrowing it
Nothing to install: this file belongs to melgarafael/DeskcommCRM. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/melgarafael/DeskcommCRM/main/.agents/skills/deskcomm-prompt/SKILL.mdgit clone --depth 1 https://github.com/melgarafael/DeskcommCRMWrote 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/melgarafael/deskcommcrm/deskcomm-prompt)<a href="https://agentmods.dev/skills/melgarafael/deskcommcrm/deskcomm-prompt"><img src="https://agentmods.dev/badge/skills/melgarafael/deskcommcrm/deskcomm-prompt/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/melgarafael/deskcommcrm/deskcomm-prompt"><img src="https://agentmods.dev/badge/skills/melgarafael/deskcommcrm/deskcomm-prompt.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 | $0.00158 | $0.01506 |
| Opus 5.5 | $0.00063 | $0.00602 |
| Sonnet 5.5 | $0.00032 | $0.00301 |
| Haiku 4.5 | $0.00016 | $0.00151 |
Grade A, and why
deskcomm-prompt 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 1mo 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Afinar o prompt de um agente com dados
"O agente está ruim" quase nunca é o prompt inteiro. Às vezes é o roteador mandando para o agente errado; às vezes é um portão de saída vetando por vocabulário que o próprio prompt ensinou; às vezes é a base de conhecimento que não cobre; às vezes é o horário. Otimizar prompt sem olhar os dados troca um problema por outro. Este guia diagnostica primeiro.
Como você age
- Sabe o que o prompt controla e o que não controla. O texto do agente é uma camada dentro
de um prompt montado pelo motor, e onze portões barram a resposta na saída. Reescrever regra
que o motor já impõe é ruído; escrever jargão ou "encaminhe ao gerente" cria veto e mata a
agenda. Lista completa em
references/anatomia-e-antipadroes.md. - Diagnóstico antes da reescrita. Execuções, vetos, handoffs, consultas à base, custo por
turno:
references/diagnostico.md. Só depois disso você abre o prompt. - Amostra mínima de conversa, com ciência da pessoa. Ler mensagens é ler dado pessoal de
terceiros. Quando for indispensável, poucas conversas, só o trecho, só na instalação, e a
pessoa sabe. Preferir agregados (o guia
deskcomm-metricascobre). - Versão nova, nunca edição. A otimização vira uma versão em rascunho, testada pelo botão Testar com o mesmo roteiro antes e depois; o clique em Publicar é da pessoa. Versão publicada é imutável e reversível.
- Uma mudança por vez, medida. Trocar cinco coisas e a conversão subir não ensina nada.
Passo 0 — o que a pessoa vê
Uma pergunta por vez: qual agente; o que acontece (a frase exata que incomoda, se houver); desde quando; o que mudou perto disso (versão publicada, materiais, roteador, número novo). Descubra a versão publicada e a data — as métricas comparam "antes/depois" por essa data.
Passo 1 — o diagnóstico
Rode o bloco de references/diagnostico.md: vetos por portão nas execuções recentes (se o veto é
internal_vocabulary ou promise, a origem costuma ser o prompt); handoffs e casos (passa demais
ou de menos); consultas à base sem acerto (material que falta × limiar); roteador errando
(intenções, não prompt); custo e tamanho do turno; fora da janela. Cada sinal aponta para uma
causa — e nem toda causa é prompt.
What ships with it
3 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.
- 1mo ago Changed d7a71fb1361c
- 1mo ago First seen · 91 lines · 158 tokens per session scan A 887106679f6d
deskcomm-prompt is a skill published in the GitHub repository melgarafael/DeskcommCRM (5,198 stars, last pushed today), licensed MIT. It adds 158 tokens to every session and 1,506 once invoked, about $0.0006 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 2026-09-12.
deskcomm-prompt compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| deskcomm-promptmelgarafael/DeskcommCRM | Skill | 5.2k | today | MIT | A | 158 | Codex |
| dspyNousResearch/hermes-agent | Skill | 261k | today | MIT | A | 19 | Claude Code, Codex |
| ai-engineering-toolkitsickn33/agentic-awesome-skills | Skill | 49k | yesterday | MIT | A | 47 | Claude Code |
| llm-app-patternsdavila7/claude-code-templates | Skill | 33k | today | MIT | A | 54 | Claude Code, Codex |
| ai-productsickn33/agentic-awesome-skills | Skill | 49k | yesterday | MIT | A | 22 | Claude Code |
| llm-opsdavila7/claude-code-templates | Skill | 33k | today | MIT | A | 44 | Claude Code, Codex |
| llm-securityzhaoxuya520/reverse-skill | Skill | 44k | 1mo ago | MIT | A | 37 | Claude Code, Codex |