Unpopular opinion: big skill packs make Claude Code worse for client work. Every extra skill is another chance for the wrong one to trigger, and none of them know which client you're working on. So I went the other way and kept only the 5 procedures I actually use every week, and open-sourced them: 1. client-workspace: one folder per client with its own rules, voice and approved claims, so Client A's tone never ends up in Client B's email 2. scope-guard: checks every "can you just quickly..." against the signed scope, quotes the extra, and drafts a polite reply (offers a paid add-on or a swap) 3. quick-quote: prices a project from 6 numbers about the client's workflow instead of your hours, and tells you when you're underquoting 4. offer-builder: turns "I do AI automation" into a fixed-scope offer with a price ladder 5. cold-email-agency: 4-step outreach sequences for small business owners There's also a full worked example (a fictional plumbing company) so you can see a real scope check end to end: 3 asks in one client email, one in scope, two quoted, reply drafted. MIT licensed, no email needed: https://github.com/CalebAnderson-Coder/ai-agency-os-starter If you try it, I'd like to know what breaks or what you'd cut.

Been building this for a while and it's finally out. It's a Claude Code setup for running several agency clients without their stuff bleeding into each other. What's in it: - a separate workspace per client (voice, claims, scope) - 14 procedures and 5 agents - a reviewer agent that checks the work before it reaches you - automation that only writes drafts, so nothing goes out without you approving it Honest note: so far I've only tested the skills on fictional client setups. If you run it on a real client I'd like to hear what breaks. There's a free starter if you want to try it first. Link's in the comments.

Most AI automation agencies pitch real estate clients the same thing: "We'll automate your lead capture." The client nods. Signs. Then 3 weeks later asks why their pipeline is full but nothing is closing. Here's the problem: lead capture is Layer 1. Agencies stop there. The real value — and the reason clients keep paying — is in Layers 2 and 3. After building automation systems specifically for real estate, here's the framework that actually moves the needle: --- LAYER 1 — Capture & Qualify (where most agencies live) - AI chatbot on website/WhatsApp captures inbound leads - Qualification flow: budget, timeline, location, type of property - Unqualified leads → nurture sequence - Qualified leads → CRM with full context This is table stakes. Every agency can build this. It's not your moat. --- LAYER 2 — Speed-to-Context (where deals are won or lost) Real estate runs on speed AND relevance. A lead that fills out a form at 11pm doesn't want a generic "thanks for reaching out" at 9am. What actually works: - Instant AI response that mirrors the lead's exact words back ("You mentioned you're looking for a 3BR in [area] under $X — here's what we have right now") - Auto-enrichment: pull active listings matching their criteria before the agent even sees the lead - Brief the agent in real time — not a raw form submission, but a structured summary: lead score, intent signals, best time to call, suggested opening line The agent walks into every call already knowing more than the lead expects. That's the unfair advantage. --- LAYER 3 — Pipeline Intelligence (where retention lives) This is what makes clients never want to leave you: - Track where every lead stalls in the pipeline (most stall at "follow-up #2" — agents give up too early) - Trigger re-engagement sequences when leads go cold (based on days since last touch, not manual reminders) - Weekly AI-generated pipeline digest for the agent: who to call today, who's about to go cold, who just re-engaged