DeskcommCRM: Skill for Codex

.agents/skills/deskcomm-metricas/SKILL.md

deskcomm-metricas is a skill for Codex from melgarafael/DeskcommCRM. It costs 166 tokens per session (1,731 once invoked), scanned A, original, MIT.

A data-analysis guide for DeskcommCRM installations. It defines how to measure conversion, sales stages, response time, handoffs, follow-ups, AI cost, losses, and busy periods using privacy-preserving database summaries.

In plain words
What is it for?
Use it to answer questions about performance, conversion, sales funnels, AI spending, handoffs, follow-ups, and the times when an operation is busiest.
Why use it?
It prevents misleading comparisons caused by different screens using different definitions, dates, or time zones. It also avoids sending personal message content to the analysis.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: installed under .agents/ (shared by several agents).

This is melgarafael/DeskcommCRM's own configuration. It tells Codex how to work on DeskcommCRM itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything DeskcommCRM configures →

About the project

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.

melgarafael/DeskcommCRM · 5,198 stars · on GitHub · deskcomm.com.br

Reuse

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.

Copy the file
curl -O https://raw.githubusercontent.com/melgarafael/DeskcommCRM/main/.agents/skills/deskcomm-metricas/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/melgarafael/DeskcommCRM

Made for: Codex.

Wrote 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.

agentmods badge for deskcomm-metricas

README.md
[![agentmods](https://agentmods.dev/badge/skills/melgarafael/deskcommcrm/deskcomm-metricas/github.svg)](https://agentmods.dev/skills/melgarafael/deskcommcrm/deskcomm-metricas)
Your own site
<a href="https://agentmods.dev/skills/melgarafael/deskcommcrm/deskcomm-metricas"><img src="https://agentmods.dev/badge/skills/melgarafael/deskcommcrm/deskcomm-metricas/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.

agentmods 80×15 button for deskcomm-metricas

Your own site · 80×15
<a href="https://agentmods.dev/skills/melgarafael/deskcommcrm/deskcomm-metricas"><img src="https://agentmods.dev/badge/skills/melgarafael/deskcommcrm/deskcomm-metricas.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 166 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,731 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00166 $0.01731
Opus 5.5 $0.00066 $0.00692
Sonnet 5.5 $0.00033 $0.00346
Haiku 4.5 $0.00017 $0.00173

Measured 1mo ago against content hash b7c27e96b438, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

deskcomm-metricas 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.

.agents/skills/deskcomm-metricas/SKILL.md · 98 lines

How it starts

The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Analisar as métricas de uma instalação

O produto mostra números em várias telas — Desempenho, Radar de risco, Uso e orçamento, Evolução da IA, Execuções, Atividades — e cada uma usa a própria régua: "handoff" tem três definições diferentes em três telas; "ganho" é crm_leads numa e a transição do agente noutra; um dia é UTC numa tela e o fuso da organização em outra. Um analista que soma números de telas diferentes chega a conclusões erradas com cara de precisas. Este guia lê a fonte (o banco), declara a régua ao lado de cada número e transforma medida em decisão.

Como você age

  • Pergunta de negócio primeiro, consulta depois. "A conversão caiu?" vira: qual funil, qual período, comparado a quê, medido por qual definição. Sem isso, o número não responde nada.
  • Agregados, nunca o texto. Você lê contagens, medianas, taxas e enums. Nunca o corpo de mensagem, nome, telefone, notas do agente, motivo de perda em texto livre sem tratar. A instalação é de terceiros e a LGPD é nativa: o que sai do banco para o modelo é só o que a anonimização preserva (ids, datas, estados, valores). references/acesso-e-lgpd.md.
  • Régua ao lado do número. Todo número vem com: período fechado [de, até), fuso, definição (qual tabela, qual filtro), e o que ficou de fora (grupos, arquivados, custo sem preço).
  • null não é zero. "Sem dado" e "zero" são respostas diferentes; o produto trata assim e você também.
  • Compare com uma referência. Período anterior, outro agente, outro funil, ou a meta que a pessoa disse. Número solto não é análise.
  • Termine em ação. Cada achado vira uma hipótese e um próximo passo: ajustar prompt (guia deskcomm-prompt), follow-up, horário do agente, base de conhecimento, equipe, ou "coletar mais 2 semanas".

Passo 0 — onde está o dado e como chegar nele

Não há banco na VPS: o Postgres é o Supabase, alcançado pela connection string SUPABASE_DB_URL do .env da instalação (Session pooler). Não existe papel de banco só-leitura no produto — a string do app enxerga todas as organizações da instalação, então toda consulta filtra organization_id. Como obter, o que pedir à pessoa, o que fazer com uma instalação de várias organizações e a alternativa por MCP: references/acesso-e-lgpd.md.

Read the full file on GitHub · 98 lines

Files

What ships with it

4 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.

Changes

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.

  1. 1mo ago Changed b7c27e96b438
  2. 1mo ago First seen · 98 lines · 166 tokens per session scan A 2dbdd0083b55

Subscribe to this mod's changes

deskcomm-metricas is a skill published in the GitHub repository melgarafael/DeskcommCRM (5,198 stars, last pushed today), licensed MIT. It adds 166 tokens to every session and 1,731 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 2026-09-12.

Related

deskcomm-metricas compared with skills from other repositories

deskcomm-metricas next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
deskcomm-metricasmelgarafael/DeskcommCRMSkill5.2ktodayMITA166Codex
rag-knowledgevstorm-co/full-stack-ai-agent-templateSkill2.0k7d agoMITA76Claude Code
fastgpt-docker-deploylabring/FastGPTSkill30kyesterdaynoneA84Claude Code, Codex
browser-f8f2966eFuRongJun-1999/dsh-memorySkill604todayMITA72Claude Code
handoff-linksHybridAIOne/hybridclawSkill159yesterdayMITA82Claude Code
stripeHybridAIOne/hybridclawSkill159yesterdayMITA23Claude Code
postgres-semantic-searchlaguagu/claude-code-nextjs-skillsSkill724d agoMITA84Claude Code