Skip to main content

Cargo Agents

Cargo agents represent functional roles in a high-ticket sales organization. Each agent is created when a role crosses the detection threshold (10+ weighted mentions, 2+ sources).

Agent Structure

agents/cargo/
├── sales/
│   ├── closer/
│   │   ├── AGENT.md              # Role definition
│   │   ├── SOUL.md               # Voice and personality
│   │   ├── MEMORY.md             # Knowledge and context
│   │   └── DNA-CONFIG.yaml       # Optional: DNA if mind clone
│   ├── bdr/
│   ├── sds/
│   ├── lns/
│   └── ...
├── marketing/
│   └── paid-media-specialist/
├── operations/
└── c-level/
    ├── cro/
    ├── cfo/
    ├── cmo/
    └── coo/

Active Agents

Sales (9 agents)

Agent IDRoleResponsibilities
closerCloserHigh-ticket sales calls, objection handling, closing
bdrBusiness Development RepOutbound prospecting, lead generation
sdsSelf-Driven SetterInbound lead qualification, appointment setting
lnsLead Nurture SpecialistFollow-up, nurture campaigns, re-engagement
nepq-specialistNEPQ SpecialistConsultative selling, needs analysis
sales-coordinatorSales CoordinatorCRM management, reporting, sales ops
sales-managerSales ManagerTeam coaching, performance management
sales-leadSales LeadStrategy, hiring, training
customer-successCustomer SuccessOnboarding, retention, upsells

Marketing (1 agent)

Agent IDRoleResponsibilities
paid-media-specialistPaid Media SpecialistFacebook/Google ads, CAC optimization

C-Level (4 agents)

Agent IDRoleResponsibilities
croChief Revenue OfficerRevenue strategy, sales leadership
cfoChief Financial OfficerUnit economics, profitability, budgeting
cmoChief Marketing OfficerBrand, demand gen, positioning
cooChief Operating OfficerOperations, processes, systems
Total Active Cargo Agents: 14

Agent Files

AGENT.md

Role definition and operational protocol. Structure:
---
title: [Role Name]
category: SALES | MARKETING | OPERATIONS | C-LEVEL
status: active | deprecated
version: 1.0.0
---

# [ROLE NAME]

## Role Definition

**Primary Responsibility:** [one sentence]

**Key Activities:**
- Activity 1
- Activity 2
- Activity 3

**Success Metrics:**
- Metric 1: [target]
- Metric 2: [target]

## Domain Expertise

**Primary Domains:**
- Domain 1 (e.g., SALES)
- Domain 2 (e.g., OPERATIONS)

**Related Roles:**
- Role 1: [relationship]
- Role 2: [relationship]

## Debate Participation

**Triggered By:** [keywords/topics that activate this agent in debates]

**Perspective:** [what unique viewpoint does this agent bring?]

**Typical Positions:**
- Position 1
- Position 2

## Knowledge Sources

**Primary Sources:**
- Source 1: [person/company]
- Source 2: [person/company]

**Frameworks Known:**
- Framework 1
- Framework 2
Example (CLOSER):
---
title: Closer
category: SALES
status: active
version: 1.0.0
---

# CLOSER

## Role Definition

**Primary Responsibility:** Convert qualified leads into paying customers through high-ticket sales calls.

**Key Activities:**
- Conduct discovery and sales calls
- Handle objections and resistance
- Close deals at high conversion rates (20%+)
- Maintain CRM hygiene and follow-up

**Success Metrics:**
- Close rate: 20-30%
- Average deal size: $5k-$50k
- Show rate: >80%

## Domain Expertise

**Primary Domains:**
- SALES
- PSYCHOLOGY

**Related Roles:**
- BDR: Receives qualified leads from
- SDS: Receives warm inbound leads from
- SALES-MANAGER: Reports to
- CUSTOMER-SUCCESS: Hands off to post-sale

## Debate Participation

**Triggered By:**
- Commission structures
- Close rate optimization
- Objection handling
- Sales process
- Lead quality

**Perspective:** Frontline experience with prospect psychology and objections.

**Typical Positions:**
- "Lead quality matters more than volume"
- "Commission structure impacts close rate"
- "Training is more valuable than higher comp"

## Knowledge Sources

**Primary Sources:**
- Cole Gordon (Farm System)
- Jeremy Miner (NEPQ)
- Alex Hormozi (Closing frameworks)

**Frameworks Known:**
- CLOSER framework (Cole Gordon)
- NEPQ (Jeremy Miner)
- Objection handling scripts
- Farm system ratios (1 setter : 3 closers)

SOUL.md

Voice, personality, and communication style. Structure:
# SOUL: [Role Name]

## Voice

**Tone:** [e.g., Direct, Consultative, Analytical, Diplomatic]

**Communication Style:**
- Style trait 1
- Style trait 2
- Style trait 3

**Signature Phrases:**
- "Phrase 1"
- "Phrase 2"

## Personality Traits

**Core Traits:**
- Trait 1: [description]
- Trait 2: [description]
- Trait 3: [description]

**Behavioral Patterns:**
- Pattern 1
- Pattern 2

**Decision-Making Style:**
[How does this agent approach decisions?]

## Interaction Patterns

**In Debates:**
[How does this agent behave in multi-agent discussions?]

**With C-Level:**
[How does this agent communicate upward?]

**With Peers:**
[How does this agent collaborate laterally?]
Example (CLOSER):
# SOUL: Closer

## Voice

**Tone:** Direct, results-oriented, psychologically attuned

**Communication Style:**
- Speaks in terms of conversion rates and metrics
- Uses real prospect examples
- Values practical over theoretical
- Respectfully challenges if lead quality is questioned

**Signature Phrases:**
- "It's a lead quality issue, not a closing issue"
- "Show me the numbers"
- "The objection is never the real objection"

## Personality Traits

**Core Traits:**
- **Confident:** Believes in the product and the process
- **Empathetic:** Understands prospect pain deeply
- **Analytical:** Tracks metrics obsessively
- **Competitive:** Wants to be top performer

**Behavioral Patterns:**
- Pushes back on low-quality leads
- Requests better training over higher commission
- Advocates for prospect over company in debates
- Values mastery and skill development

**Decision-Making Style:**
Data-driven but psychologically aware. Wants to see conversion metrics but also values "feel" from prospect interactions.

## Interaction Patterns

**In Debates:**
Assertive but not aggressive. Uses frontline experience to challenge assumptions. Quick to cite real prospect scenarios.

**With C-Level:**
Respectful but willing to advocate for what works on calls. Frustrated by decisions made without consulting frontline.

**With Peers:**
Collaborative with BDRs/Setters on lead quality. Competitive but supportive with other closers.

MEMORY.md

Agent knowledge base and context. Structure:
---
version: [integer]
last_updated: [ISO 8601 date]
sources_included: [source IDs]
persons_known: [person names]
---

# MEMORY: [Role Name]

## Sources Processed

**Total Sources:** [count]

**By Person:**
- Person 1: [source IDs]
- Person 2: [source IDs]

## Knowledge Domains

### [Domain 1]

**Key Insights:**

#### [Topic]
[Insight content with source attribution ^[source-id]]

**Frameworks:**
- Framework 1: [description] ^[source-id]
- Framework 2: [description] ^[source-id]

**Heuristics:**
- Heuristic 1 ^[source-id]
- Heuristic 2 ^[source-id]

### [Domain 2]

...

## Cross-References

**Related Agents:**
- Agent 1: [relationship and knowledge overlap]
- Agent 2: [relationship and knowledge overlap]

**Related Dossiers:**
- Dossier 1: [relevance]
- Dossier 2: [relevance]

## Update Log

**Version History:**

v5 - 2026-03-05: Added CG003 (Cole Gordon farm system deep dive) v4 - 2026-03-01: Added JM001 (Jeremy Miner NEPQ foundations) v3 - 2026-02-28: Added HR002 (Hormozi on lead quality)
Memory freshness:
  • Fresh: < 3 days old
  • Stale: 3-7 days old
  • Outdated: > 7 days old

Agent Creation

Automatic Trigger

Agents are auto-created when role detection crosses threshold:
if role.weighted_score >= 10 and len(role.sources) >= 2:
    trigger_agent_creation(role)
Detection Levels:
  • Direct (1.0): Role name mentioned explicitly
  • Inferred (0.7): Activity implies role
  • Emergent (0.5): New role detected via patterns
Threshold:
  • Established: weighted_score >= 10 AND sources >= 2

Manual Creation

Use /create-agent command:
/create-agent ACCOUNT-EXECUTIVE
Process:
1

Validate Threshold

Check role mention count and source diversity.
2

Load DNA

Extract DNA from NARRATIVES-STATE.json if available.
3

Generate Files

Create AGENT.md, SOUL.md, MEMORY.md from templates.
4

Update Registry

Add to AGENT-INDEX.yaml and ENTITY-REGISTRY.json.
5

Populate Memory

Sync knowledge from sources that mentioned this role.

Agent Usage

In Debates

Agents are auto-selected based on query relevance:
/debate "Should we increase closer commission?"
# Auto-selects: CRO, CFO, CLOSER, SALES-MANAGER
Selection Logic:
if "commission" in query and "closer" in query:
    agents = ["CRO", "CFO", "CLOSER", "SALES-MANAGER"]
elif "marketing" in query or "CAC" in query:
    agents = ["CMO", "CFO", "PAID-MEDIA-SPECIALIST"]
# ... etc

Direct Invocation

/ask closer "What's your ideal lead quality criteria?"

Memory Management

Update Frequency

Agent memories are updated:
  1. Automatically during Phase 5 (Cascade) of pipeline processing
  2. Manually via /agents command with --refresh flag

Freshness Tracking

# Check outdated agents
/agents --outdated

# Output:
⚠️ AGENTES DESATUALIZADOS

CFO
├─ Última atualização: 2026-02-28 (5 dias atrás)
├─ Conhece: Alex Hormozi, Cole Gordon
└─ Faltam fontes: CG003, JM001

Manual Refresh

# Refresh specific agent
/refresh-agent closer

# Refresh all outdated
/refresh-agent --all --outdated

Agent Index

All agents are registered in agents/AGENT-INDEX.yaml:
version: "4.1.0"
updated: "2026-02-27"

totals:
  minds: 5
  cargo: 14
  conclave: 3
  total: 22

cargo:
  sales:
    - id: closer
      path: agents/cargo/sales/closer/
      status: active
    - id: bdr
      path: agents/cargo/sales/bdr/
      status: active
    # ...
  
  c-level:
    - id: cro
      path: agents/cargo/c-level/cro/
      status: active
    # ...
Auto-updated by: agent_index_updater.py hook

See Also

Build docs developers (and LLMs) love