Learnings - Concepts - go-to-market

Swan GTM skills architecture

Swan's GTM skills architecture splits agent knowledge into foundations and maps. Foundations hold reusable GTM judgment. Maps are thin workflow routes that open the right foundations in the right order.

Maps 17 workflowsFoundations 7 truthsPattern how vs where
GTM event->map route+foundation context->agent action
00

The architecture at a glance

The useful lesson is not the exact Swan playbook. It is the separation of concerns: foundations own reusable judgment, while maps own workflow order. When ICP, scoring, routing, or voice changes, the update happens once in a foundation and every map that reads it benefits.

7
Foundations
CRM, Research, ICP...
17
Maps
funnel workflows
1x
Horizontal edit
change a truth once

Before: context embedded in every workflow

Workflow A: ICP + scoring + routing + voiceWorkflow B: ICP + scoring + routing + voiceWorkflow C: ICP + scoring + routing + voice

One ICP change becomes many edits, and drift is hard to spot.

After: workflows point into shared foundations

Workflow A -> ICP foundationWorkflow B -> ICP foundationWorkflow C -> ICP foundation

One foundation change updates every map that reads it.

01

Foundations hold the how

A foundation is reusable context that should not belong to any one play. Pick a foundation to see what it owns, its public sub-skills from Swan's playbook, and which maps read it.

Foundation

CRM

Lookup, dedupe, ownership, deal state, data quality, and write permissions.

/lookup/dedupe/Deal Management/Data Quality
Owns

How agents read, trust, update, and avoid duplicating CRM records.

AgencyCore lesson

Every agent that touches a person, company, deal, or signal should use one shared CRM contract.

Maps that read it
02

Maps hold the where

A map is one workflow written as a route. It carries little logic of its own. It declares the trigger, then opens the foundations needed for that GTM moment.

MidCRM event

Inbound demo

Check whether the account is known, brief the rep, qualify, score, route, draft, and alert.

MAPWhat this map teaches

High-intent inbound should route through every shared judgment layer.

03

Inbound demo as a route

Swan's public playbook shows inbound demo as the clearest map: seven stops, each opening a foundation sub-skill. Click a stop to see what the agent is loading at that moment.

Stop 01

CRM /lookup

Do we know this account?

This is progressive disclosure for GTM context. The agent does not carry the whole GTM brain at once. The map loads only the sub-skill needed for the current step.
04

Where logic should live

The rule is strict: if the logic repeats across workflows, it belongs in a foundation. If it only defines one workflow's order or trigger, it belongs in a map.

Foundation

Who is good?

ICP fit, exclusion rules, segment caveats, and buyer roles belong in reusable context.

05

Borrowing this for AgencyCore

We can adapt the pattern into our own agent context architecture: shared foundations for GTM judgment, thin maps for workflows, and evals around both.

Step 01

Inventory duplicated context

Find where ICP, scoring, routing, voice, and alert payloads are repeated across AgencyCore workflows.

OutputA duplication map

Good first foundation

Start with CRM because every GTM map needs to know whether a person, company, deal, or signal already exists. That gives agents a shared read/write contract before they act.

Good first map

Use an inbound lead or visitor reveal map first. The trigger is concrete, the route is short, and the human review gate can stay close to the rep.

06

Sources and caveats

This page is grounded in Swan's public post, public site, and public playbook bundle. The original post mentions Positioning, but the public architecture renders seven foundations without a separate Positioning block, so this page treats Positioning as folded into ICP, Research, or Outreach Voice.

AgencyCore - Learnings - Swan GTM skills architecturefoundations - maps - context engineering