The Council has adjourned

A few months ago I was building Lead Scout — a personal AI pipeline that finds companies adopting AI, scores them for fit, tracks down a decision-maker, and drafts me a personalised cold email. All automated. All local. No monthly SaaS bill.

It worked. Kind of. The results were noisy, the email drafts ran too long, and I had a nagging feeling that something was architecturally off but I couldn’t put my finger on what. I was patching symptoms instead of diagnosing the system.

Then I came across a post from someone in this community — I genuinely cannot remember your name and it’s driving me mad — but they were using Norse Gods as AI agent roles to review and improve their project. Thor for raw power, Odin for wisdom, that kind of thing. Whoever you are: thank you. That post changed how I think about working with AI. If anyone knows who I’m talking about, please tag them. I owe them a coffee.

The idea clicked something in my brain. What if instead of asking one Claude instance to do everything, I gave each review role its own identity, its own domain, and its own rules?

So I built The Council. Three agents. Egyptian theme, because why not.

Ra — The Architect

Ra sees the whole system from above. He doesn’t touch code. He thinks in pipelines, sequences, tool choices, and strategic gaps. Before any major change, Ra reviews the architecture and asks: are we building the right thing, in the right way, for the right reasons? Ra runs twice per session — same prompt, two separate Claude instances — because he often approaches the same problem from different angles. Concerns that appear in both sessions are high priority. Concerns that appear in only one are still worth investigating.

Anubis — The Surgeon

Anubis goes stage by stage through the actual code. File names. Function names. Exact failure conditions. He doesn’t guess and he doesn’t generalise. If Ra says “the scoring stage feels fragile,” Anubis finds exactly which line is fragile and exactly why. He only reports what he can point to specifically. He weighs everything on the scales — nothing passes without evidence.

Horus — The Watchman

Horus reads the session logs. I’ve been tracking every Lead Scout run in a spreadsheet — quality scores, noise levels, contact quality, whether the emails were sendable. Horus looks at that data across time and spots what nobody else is looking for yet: drift. A metric quietly getting worse over five sessions. An industry/region combo that’s consistently underperforming. A hidden gem nobody’s been paying attention to. By the time Anubis finds the wound, Horus has already seen the bruise forming.

The process is simple:

1. Run Ra twice (parallel Claude Code chats). Merge his strategic concerns.

2. Run Horus in parallel. He reads the session log for pattern drift.

3. Run Anubis with Ra’s concerns and Horus’s drift alerts. He pinpoints the code.

4. Return to Ra with all findings. He produces a recovery and improvement plan — maximum 10 sentences, no more.

5. Chris (me) executes. Chris always decides.

What changed after installing The Council?

The scoring layer got smarter. We added size penalties, hard rejects for certain company types, competitor detection that changes the entire email angle, and language detection so Spanish and German companies get outreach in their own language. The discovery stage got cleaner query sets per industry with negative filters to kill the noise. The email drafting layer got a proper persona-encoded system prompt with dynamic branching based on company type — client, employer, or competitor. Each gets a completely different email.

Most importantly: I stopped asking “why isn’t this working” and started asking “which stage is this problem in, and who do I send it to?”

I’m an AI Adoption Trainer. I’ve spent 30 years teaching people that scary, complex skills become manageable when you break them into clear roles with clear rules. Turns out that applies to AI agents too.

This is what I love about this community. Someone posts an idea with Norse Gods, I steal it shamelessly, theme it around ancient Egypt, and my lead generation pipeline gets measurably better. We’re all figuring this out together.

If you’re building something with AI and you’re still treating your Claude (or GPT, or whatever) as one giant do-everything assistant — try giving your reviews a character. Give them a domain they own. Give them rules about what they’re NOT allowed to do. Watch what happens.

The code for Lead Scout is open. The Council system is reusable for any pipeline. Happy to share if anyone wants it.

And seriously — if you know who posted the Norse Gods agent framework, please tag them below. They deserve the credit.

19 comments