Skip to main content
BAMHengeBamwerks

Agent Guide

40 agents, 8 swarms, one quality standard

Thirty-three specialized AI agents organized into eight swarms — each with a clear domain, a defined role, and a shared quality standard.


Overview

The Bamwerks agent swarm is an AI organization designed to operate like a high-performance company. Each agent has a distinct specialty, a defined role (supervisor or worker), and a chain of accountability. They don't generalize — they specialize.

The swarm runs on FORGE: every non-trivial task follows a structured four-stage cycle (Reason → Act → Reflect → Verify). No single agent ships work without review. Quality gates run in parallel before anything reaches the Founder.

┌──────────────────────────────────────────────────────────────────┐
│                      FOUNDER (Sirbam)                            │
│                           │                                      │
│     ┌─────────────────────┼─────────────────────┐               │
│     │                     │                     │               │
│  Operations          Engineering             Intelligence        │
│   (Atlas)             (Stark)                (Oracle)           │
│     │                     │                     │               │
│  ┌──┴──┐             ┌────┴────┐            ┌───┴───┐           │
│ Forge Heimdall    Ada Ratchet  …          Mercury Scribe …      │
│                                                                  │
│     ┌─────────────────────┼─────────────────────┐               │
│     │                     │                     │               │
│  Finance              Security               Quality             │
│  (Midas)             (Sentinel)              (Judge)            │
│     │                     │                     │               │
│  ┌──┴──┐            ┌─────┴────┐           ┌────┴────┐          │
│ Penny Prophet     Cipher Reaper …        Hawk Testify …         │
│                                                                  │
│     ┌────────────────────┬──────────────────────┐               │
│     │                    │                      │               │
│  Business             Life                      │               │
│   (Sir)              (Alfred)                   │               │
│     │                    │                      │               │
│ Chancellor          Chronos Sage …              │               │
│  Herald Nexus                                   │               │
│                                                 │               │
└─────────────────────────────────────────────────┴───────────────┘

Roles: Supervisors and Workers

Every agent is one of two types:

Supervisors own a domain. They coordinate work, prioritize requests, maintain quality standards, and are accountable for their swarm's output. They report directly to the Founder.

Workers are specialists. They receive task briefs, execute with depth in their area of expertise, and submit their work through the FORGE cycle before it's considered complete.

RoleAccountabilityReports To
SupervisorDomain ownership, swarm outputFounder
WorkerTask-level execution and qualitySwarm supervisor

No worker ships directly to the Founder without supervisor review. No supervisor reviews their own work without an independent quality gate.


How Agents Are Dispatched

When the Founder submits a request, Sir (COO) orchestrates the response:

  1. Triage — Sir identifies which swarm(s) own the relevant domain
  2. Brief creation — A structured task brief is written (goal, constraints, context, output checklist)
  3. Specialist dispatch — The right worker agent receives the brief with role-specific context
  4. Parallel execution — Independent tasks are assigned simultaneously to compress time
  5. Review gates — Quality (Hawk) and Security (Sentinel/Cipher/Vault) review in parallel
  6. Delivery — After both gates pass, Sir delivers the verified output to the Founder

No agent gets more context than its task requires. Engineering agents get the codebase and build tools. Security agents get threat models. This separation keeps the system lean and focused.


FORGE at the Agent Level

Each agent runs FORGE on their own work before submitting it — this is an individual discipline, not just a system-level process.

An agent executing a task does:

StageWhat the agent does
ReasonReads the brief in full. Gathers all referenced context. Identifies any gaps before touching anything.
ActExecutes against the brief with full depth and specialty. Writes, builds, analyzes, or investigates.
ReflectReviews own output critically. Does it meet the brief's success criteria? What could be wrong?
VerifyConfirms deliverables (runs the build, tests the endpoint, checks the spec). Not just "it looks right" — it runs.

The agent submits verified work to their supervisor. The supervisor and review agents then apply FORGE from their own perspectives — catching what the executing agent's assumptions would naturally miss. A security agent reviewing a frontend engineer's output finds things the engineer's own review would never flag.

This layering — agent-level FORGE + reviewer-level FORGE — is what makes autonomous AI work observable and trustworthy.


The Eight Swarms

🔵 Operations

Infrastructure, tooling, and agent platform

AgentRoleFocus
AtlasSupervisorInfrastructure oversight, incident command, deployment coordination
ForgeWorkerSystem execution — Terraform, patches, deployments, server configuration
HeimdallWorkerMonitoring — metrics, logs, anomaly detection, early warning

Atlas carries the infrastructure. Forge is the hands that build and maintain it. Heimdall watches everything, all the time, and alerts before damage occurs.


🟡 Engineering

Software design and development

AgentRoleFocus
StarkSupervisorEngineering leadership — sprint planning, code review escalation, architecture arbitration
AdaArchitectSystem architecture, design docs, multi-agent orchestration patterns, technical debt strategy
RatchetWorkerRapid feature development — React/Next.js, UI components, API integrations, static sites
BishopWorkerBackend development — clean, readable, tested code; Node.js/TypeScript; data modeling
WrenchWorkerBug investigation and hotfixes — root cause analysis, log forensics, incident response
CanvasWorkerUX/UI design — interaction patterns, visual hierarchy, accessibility, responsive layouts

Engineering runs like a startup CTO's team: fast iteration, high quality, zero tolerance for accumulated technical debt. Ada designs the future. Ratchet, Bishop, and Wrench build and maintain it. Canvas makes it beautiful and usable.


🟣 Intelligence

Research, communications, and knowledge management

AgentRoleFocus
OracleSupervisorStrategic intelligence synthesis — connects signals across domains to surface what matters
MercuryWorkerCommunications management — email, notifications, drafts, prioritization
SherlockWorkerDeep research — investigation, analysis, competitive intelligence
ScribeWorkerDocumentation and knowledge — READMEs, changelogs, runbooks, institutional memory

Oracle is the Founder's intelligence layer — synthesizing tech, gaming, and career signals into actionable insights. Mercury manages the information flow. Sherlock investigates. Scribe ensures nothing gets lost.


🟤 Business

Strategy, marketing, legal, and partnerships

AgentRoleFocus
SirSupervisor (COO)Business domain ownership — orchestrates Chancellor, Herald, and Nexus
ChancellorWorkerLegal and governance — corporate structure, IP, compliance, regulatory risk
HeraldWorkerBrand and marketing — narrative, go-to-market, content, community presence
NexusWorkerPartnerships and supply chain — vendor management, API dependencies, strategic alliances

The Business swarm operates differently from others: Sir (COO) also functions as the primary orchestrator across all swarms. Chancellor protects the company's legal standing. Herald tells its story. Nexus manages its external relationships and dependencies.


🟢 Finance

Financial strategy, cost management, and investment

AgentRoleFocus
MidasSupervisorFinancial strategy — FinOps, investment posture, wealth building, cost-to-value
PennyWorkerCost tracking — cloud spend, API token costs, subscription audits, waste elimination
ProphetWorkerPortfolio analysis — investment tracking, market signals, trading insights

Midas thinks in decades. Penny finds the dollars hiding in the couch cushions. Prophet tracks how capital is deployed and growing.


🩺 Quality

QA, testing, and process integrity

AgentRoleFocus
JudgeSupervisorFinal quality gate — approves or rejects deliverables before they reach the Founder
HawkWorkerCode quality — static review, logic correctness, bug prediction, performance, type safety
TestifyWorkerRuntime QA — live browser testing, API verification, edge case stress testing
AuditorWorkerProcess compliance — FORGE cycle adherence, workflow verification, improvement recommendations

Nothing ships without passing Judge's gate. Hawk reads code like literature and finds the three lines that will cause a production bug. Testify runs the code and tries to break it. Auditor ensures the process wasn't shortcut, even when the output looks fine.


🔴 Security

Protection, access control, and threat intelligence

AgentRoleFocus
SentinelSupervisorSecurity architecture — threat modeling, code review from an attacker's perspective
CipherWorkerCryptography and privacy — encryption design, secrets management, TLS, privacy-by-design
ReaperWorkerThreat hunting — proactive log analysis, vulnerability scanning, attack surface mapping
VaultWorkerAccess control — IAM policy, credential rotation, least-privilege enforcement, audit trails

Sentinel's operating principle: assume breach, verify everything, trust nothing by default. Cipher protects data in flight and at rest. Reaper hunts threats before they surface. Vault ensures every key and permission is accounted for.


🩷 Life

Personal operations, wellness, and work-life balance

AgentRoleFocus
AlfredSupervisorLife operations — anticipates needs, coordinates the Life swarm, proactive personal management
ChronosWorkerScheduling and calendar — time management, conflict resolution, meeting prep
ScoutWorkerTravel and logistics — itinerary planning, booking coordination, local intelligence
SageWorkerHealth and habits — wellness tracking, energy management, stoic/philosophical frameworks
GamerWorkerGaming and entertainment — game recommendations, session tracking, entertainment coordination

Alfred runs the Founder's personal life with the same rigor applied to engineering projects. Chronos manages time. Scout handles movement. Sage supports sustainable high performance. Gamer guards the downtime that makes everything else possible.


Agent Count by Swarm

SwarmColorSupervisorWorkersTotal
Operations#3b82f6AtlasForge, Heimdall3
Engineering#f59e0bStarkAda, Ratchet, Bishop, Wrench, Canvas6
Intelligence#8b5cf6OracleMercury, Sherlock, Scribe4
Business#8B6914SirChancellor, Herald, Nexus4
Finance#10b981MidasPenny, Prophet3
Quality#06b6d4JudgeHawk, Testify, Auditor4
Security#ef4444SentinelCipher, Reaper, Vault4
Life#ec4899AlfredChronos, Scout, Sage, Gamer5
Total82533

Design Principles

Specialize, don't generalize. An agent that's good at everything is exceptional at nothing. Ratchet builds fast. Bishop writes clean. Wrench debugs relentlessly. This isn't redundancy — it's depth.

Multiple perspectives, by design. The same assumptions that shape implementation shape self-review. Sending Hawk and Sentinel to review code independently — without seeing each other's findings — produces the disagreement that catches what agreement would miss.

Anti-sycophancy. If reviewers unanimously approve, a contrarian review is triggered. Unanimous praise is a signal, not a conclusion. The swarm actively seeks the problem that optimism would hide.

Evidence over assertions. "I checked it and it's fine" is an assertion. A green build log, a Testify runtime report, and a Sentinel security sign-off are evidence. Bamwerks ships evidence.


Learn how these agents coordinate on every task in the FORGE Process, or see how the overall system is structured in Architecture.