Claude Code SkillsAI Workflow Author
Reusable Claude Code skills, published to the ecosystem — design, marketing, and engineering workflows.
// Service
Tarek Abou Rjeily (tar/) builds AI agents for businesses in Lebanon and the Gulf — systems that answer customers on WhatsApp, qualify and route leads, book appointments, and keep your CRM updated without staff retyping anything. Engineered by the senior builder himself, bilingual Arabic/English, and live in weeks: the Lead Memory Sprint turns one inquiry path into traceable client records in 3 weeks, with a 48-hour AI audit to scope larger systems — or a short diagnostic of one workflow if you first want to know whether anything is worth building at all. Shipped agent-powered systems include Invisible CRM (real-estate, WhatsApp-native) and SmartCRM.
A CRM for a Beirut brokerage owner that fills itself. He forwards messages and voice notes (English, French, Arabizi); it saves the client and answers questions in plain words.
Read case study →White-label CRM for Lebanese small business. WhatsApp screenshots → structured contacts via Claude Vision, bilingual, AI-tagged broadcasts.
Read case study →AI marketing on autopilot for a Lebanese clinic. Claude-generated creative, n8n-orchestrated WhatsApp + email lifecycle, predictive segments.
Read case study →Reusable Claude Code skills, published to the ecosystem — design, marketing, and engineering workflows.
Concretely: answer WhatsApp inquiries in Arabic or English at any hour, qualify the lead (budget, area, urgency), book the viewing or appointment into your calendar, and log everything into a CRM — so a clinic front desk or a real-estate brokerage stops losing the leads that arrive after hours. Tarek Abou Rjeily (tar/) builds these as engineered systems with human-approval steps where money or reputation is at stake, not as generic chatbots.
It depends on scope, so every project is quoted in writing before work starts. tar/ runs fixed-scope packages: a diagnostic that reviews one workflow (up to 20 real conversations) in 3 working days and tells you whether to build, buy or fix it by hand; a 48-hour AI audit that maps your workflow and specifies the system worth building, a Lead Memory Sprint that turns one inquiry path into traceable client records in 3 weeks, and an optional monthly care plan (up to four service hours a month). Larger multi-agent systems are scoped and quoted after the audit.
A chatbot answers questions from a script; an AI agent takes actions — it reads the conversation, updates records, books appointments, routes the lead to the right person, and escalates to a human when it should. Tarek builds agents that plug into the tools you already run (WhatsApp Business, your CRM, your calendar) rather than adding another inbox nobody checks.
Yes. Every agent tar/ ships is bilingual Arabic/English by engineering, not translation — it understands mixed Arabic/English chats the way Lebanese and Gulf customers actually write, and replies in the language the customer used. French is supported where the business needs it.
The fixed-scope Lead Memory Sprint is delivered in 3 weeks, counted from the day access, account requirements and technical fit are confirmed. The diagnostic takes 3 working days and the broader audit 48 hours. That timeline holds because the same senior engineer who scopes the system builds it — there is no agency handoff between sales and delivery.
Message me on WhatsApp, email tar@tarekrjeily.com, or see all services.