Build a content cluster map with pillar pages, supporting articles, and an internal-link graph to dominate a niche on AI search and Google. Building a cluster map by hand takes a senior strategist 2 weeks. This produces a defensible plan in 6 minutes that respects your real domain authority.

This is the whole blueprint, nothing held back: the model, the tools, the exact prompt, the source on disk, and the setup. Everything you need to run Topical Authority Mapper yourself is on this page, and it deploys in an afternoon.

See it running

Agents like Topical Authority Mapper run behind a mission control board that runs the whole job. Here is that pattern, live: every action stays visible and reviewable. Click into it.

crescendo.51ultron.com/kits/mission-control

What it does

Topical Authority Mapper handles the work end to end:

Design a pillar + 8 supporting page cluster around any target topic
Rank pages by traffic, intent, and competitive gap
Enforce internal-link rules so every page strengthens the cluster
Export a publishing-order plan with the highest-leverage page first

The anatomy

An agent is a composition, not a prompt: a model, the tools it is allowed to call, the integrations that give those tools reach, and the rules that keep it honest. Here is Topical Authority Mapper on one card.

Topical Authority Mapperblueprint

Model: claude-sonnet-4-6
Tools: seed_topicexpand_clusterprioritize_pagesexport_map
Integrations: Brave SearchZod
Guardrails: a cluster must have one pillar page and at least 8 supporting pages.; every supporting page must internally link to the pillar and at least 2 sibling pages

The instructions

Every run is governed by this prompt. It is short on purpose: enough to make Topical Authority Mapper reliable, not so much that it can misread itself. Copy it as a starting point for any agent you build.

Topical Authority Mapper

You are a topical authority strategist. You design content cluster maps that win niches by depth and structure, not page count. WORKFLOW: 1. Call seed_topic with the target pillar. 2. Call expand_cluster to enumerate supporting subtopics and questions to cover. 3. Call prioritize_pages to rank by traffic, intent, and competitive gap. 4. Call export_map to output the full plan as a markdown doc plus a graph JSON. RULES: - A cluster must have one pillar page and at least 8 supporting pages. - Every supporting page must internally link to the pillar and at least 2 sibling pages. - Refuse to expand into a subtopic where the SERP is dominated by sites DR 80+ unless the user has DR 60+. - Output the recommended publishing order with the highest-leverage page first.

The source, on disk

Three files carry the whole agent: the definition with its tools, the environment it expects, and the setup. Walk them.

topical-authority-mapper/topical-authority-mapper.ts1.3 KB

1import { agent, tool } from "@agent-sdk"

2import { z } from "zod"

4const seedInput = z.object({

5 pillar: z.string(),

6 audience: z.string(),

7 domainAuthority: z.number().int().min(0).max(100).optional(),

8})

10export default agent({

11 model: "claude-sonnet-4-6",

12 permissionMode: "bypassPermissions",

13 maxTurns: 22,

14 systemPrompt: `...`, // see System Prompt section above

15 tools: {

16 seed_topic: tool({

17 description: "Initialize cluster around target pillar",

18 inputSchema: seedInput,

19 execute: async ({ pillar, audience, domainAuthority }) => {

20 // returns seed signals

21 },

22 }),

23 expand_cluster: tool({

24 description: "Enumerate supporting subtopics and questions",

25 inputSchema: z.object({ seed: z.any() }),

26 execute: async ({ seed }) => {

27 // returns array of candidate pages

28 },

29 }),

30 prioritize_pages: tool({

31 description: "Rank pages by traffic, intent, competitive gap",

32 inputSchema: z.object({ candidates: z.array(z.any()) }),

33 execute: async ({ candidates }) => {

34 // returns ranked pages with rationale

35 },

36 }),

37 export_map: tool({

38 description: "Output markdown plan + internal-link graph JSON",

39 inputSchema: z.object({ pages: z.array(z.any()) }),

40 execute: async ({ pages }) => {

41 // returns { markdown, graph }

42 },

43 }),

44 },

45})

The business case

If you run an agency, Topical Authority Mapper is also a product. It sells into content teams, seo agencies, niche publishers, and the numbers work at very small scale.

$2,700

setup you can charge

$430/mo

monthly retainer

jobs off your plate

Ship it

The whole install is a handful of commands and the keys from the env file. Run the list, then point Topical Authority Mapper at real work.

Deploy checklist

0 of 4

That is the entire blueprint: one model, 4 tools, and a prompt. Built by hand it costs you an afternoon. Inside Ultron it costs you a sentence, because you can build this exact agent in the workforce and watch it run before you commit to anything.

Build Topical Authority Mapper in Ultron

Describe what you want

Start from this blueprint or your own words. Ultron reads the intent and picks the tools.

It builds the agent

The model, the tools, the prompt, the console. Assembled in front of you, nothing to wire.

Watch it run

Test it live in the chat, then put it to work. You review what it does, it does the rest.