How GeneWorks Works

Requirement. Design. Build. Deploy. Verify.

Six stages. Two human sign-offs. 152 specialist agents in between. This is the whole picture: how a single ServiceNow story flows through the GeneWorks delivery loop, and — just as importantly — an honest map of what the agents build end-to-end and what they won't touch.

01The continuous loop

One story. Six stages. Two places where the work stops.

Every story — a catalog tweak, a business rule edit, a scoped app feature — runs through the same six stages: Requirement, Design, Task Planner, Execute, SN Deploy, Verify. The agents drive all of them. At two points, the loop pauses for a human: design sign-off before anything is built, and verification sign-off before it ships.

1
Agent

Requirement

Requirement Agent

You talk to the Requirement agent. It turns whatever you bring — a Jira story, a CSV from a workshop, a Visio swim-lane, or plain English — into a structured, versioned Functional Requirements Spec, and asks clarifying questions where the requirement is thin. Too ambiguous to specify? It bounces back with a structured list of what's missing.

InputStory brief, CSV, Visio, plain English
OutputVersioned requirements spec (PDF)
Typical timeMinutes
2
Sign-off

Design

Design Agent → your sign-off

The Design agent produces a full Solution Design — data model, dependencies, ACL implications, integration shape, and the exact ServiceNow Fluent artifacts to build. It reads from the workspace's accumulated context, so it matches your team's conventions. Then your architect reviews and signs off — nothing builds until this happens.

InputVersioned requirement
OutputApproved solution design + artifact list
ReviewerSenior architect / platform owner
3
Agent

Task Planner

Planner Agent

The Planner agent breaks the approved design into a granular, ordered implementation plan — every table, role, group, system property, ACL, business rule, and test as its own task. You can regenerate or adjust the plan before a single line is built.

InputApproved solution design
OutputOrdered build plan (e.g. 28 tasks)
Typical timeMinutes
4
Agent

Execute

Main Agent + Specialists

The Main agent builds the tasks in parallel waves, writing real ServiceNow Fluent code (React + TypeScript) you can read in the built-in IDE. ATF and functional tests are authored alongside the build — not after the fact. Each task is marked done or flagged, and every agent sees the others' work in the same workspace.

InputOrdered build plan
OutputWorking Fluent app + tests + docs
Typical timeMinutes to hours
5
Agent

SN Deploy

Deploy

GeneWorks installs the app into your instance as a safe, non-destructive update. Any changes made on the instance since your last deploy are merged in first, so your work never overwrites theirs. Every step is checkpointed — the whole deploy is fully reversible. GeneWorks never deploys straight to production.

InputBuilt Fluent app
OutputInstalled scoped app (reversible)
Typical timeMinutes
6
Sign-off

Verify

ATF + FT → your sign-off

GeneWorks runs ATF (server + client) plus live-browser functional tests against the deployed app — on our own test runner, so you never license ServiceNow's ATF cloud runner. Failures get one-click auto-fix where the pattern is known. Then your team signs off on the evidence. No theater: if 73 of 100 stories completed clean, it says exactly that — with a handoff for the rest.

InputDeployed app
OutputSigned-off evidence + CAB-ready package
ReviewerDelivery lead + CAB
02Why the sign-offs matter

Two checkpoints. Both human. Both non-negotiable.

The gates aren't ceremony. They're the contract. Every change record points to a named human who approved the design and a named human who approved the evidence. Audit trail by construction, not by reconstruction.

Sign-off 1 · Before build

Design sign-off

The agents have specified what they will build, why, and how it touches the rest of the instance. Now a human says yes or no. No code has been written yet. Reversing course here costs minutes — reversing it after build costs days.

What gets reviewed
  • Data model changes & ACL implications
  • Dependency map & blast-radius assessment
  • Integration boundary & auth flow
  • Test coverage plan (ATF + functional)
  • Explicit "what I'm assuming" list
Sign-off 2 · Before it ships

Verify sign-off

The app is deployed and verified — ATF plus live-browser functional tests, on our own runner (no ServiceNow ATF cloud-runner license). The agents produce an evidence package per story: what passed, what's flagged, what's blocked. A human signs off before anything goes to CAB. No story is "done" because an agent says so.

What gets reviewed
  • ATF (server + client) run results
  • Live-browser functional-test results
  • Before/after configuration diffs
  • Flagged items needing human follow-up
  • CAB-ready change package
03What Gene builds

Three tiers. One straight answer per artifact type.

Most ServiceNow AI tools sell a fantasy: type a story, get production code. The reality is messier — and we'd rather you plan around it than find out at deployment. Every artifact maps to one of three tiers based on real delivery data. HIGH means the agents handle it end-to-end. MEDIUM means the agents draft it and a human reviews. LOWER means the agents assist, but a senior engineer runs the work.

High Fit
90–100% · End-to-end automation
Fully autonomous. The agent team builds, tests, evidences, and packages without a human in the middle.
Service Catalog items — variables, sets, fulfillment, approval gates
Notifications & templates — content, conditions, recipients
Client Scripts — onLoad, onChange, onSubmit, mandatory logic
UI Policies — show/hide, mandatory, read-only conditions
Knowledge Articles — KB seeding, structuring, refresh
System Properties — config flags, defaults
Scheduled Jobs — schedules, scripts, recurrence
User / Group / Role admin — onboarding, role assignment, deactivation

Reviewer effort: spot-check. The agents have a clean track record on these artifact types — review is a sanity pass.

Medium Fit
65–85% · Draft + human review
Agents produce a working draft. A senior engineer reviews logic, edge cases, and integration boundaries before merge.
Business Rules — async, before, after, query
Script Includes — server-side reusable logic
UI Actions — server & client-side execution paths
Outbound REST & Inbound Scripted REST APIs
Transform Maps — field mapping, transform scripts, coalesce logic
Service Portal Widgets — HTML, server, client controllers
Reports & Dashboards — list reports, bar/pie/trend charts
ATF Tests & Fix Scripts — generate, parameterize, run
Inbound Email Actions — parsing rules, action scripts
Scoped App scaffolding — table design, navigation, basics

Reviewer effort: meaningful. Architect reads the diff, runs the ATF, signs off. Same review as a junior dev's PR.

Lower Fit
40–65% · Assist, don't autopilot
Agents help — pattern lookup, scaffolding, code review — but a senior engineer drives the work end-to-end.
Multi-component orchestrations — workflows spanning many tables, integrations, and gates
Performance Analytics indicators — formula design, scoring, breakdowns
Service Portal full-page builds — multi-widget composition, custom theming
Mobile Studio builds — native apps, screen flows, push config

Reviewer effort: the agent is a junior pair-programmer. The architect designs and owns the work — the agent accelerates the mechanical parts.

04What we won't touch · 12 hard exclusions

Twelve areas where autonomous AI is the wrong tool.

ACL design

Security boundaries are a human-architect call. Agents don't design ACLs.

Flow Designer action graphs

Visual orchestrations remain in human hands.

CMDB Identification & Reconciliation Engine

IRE rules drive your single system of record. Don't let an agent guess.

Service Mapping & Relationship discovery

Agents don't author dependency topology.

Performance Analytics collectors

Frequency, rollup, and snapshot logic stay with the analyst.

UI Builder / Workspace experiences

Component composition is hand-crafted.

Discovery & Service Mapping configuration

Patterns, credentials, schedules — human-owned.

NLU / Virtual Agent corpus

Intent design and utterance curation is a domain task.

SSO / SAML / IdP configuration

Identity is a security boundary, not an automation target.

MID Server lifecycle

Provisioning, certificates, network paths — infra team's domain.

Production deployments

The CAB calendar exists for a reason. Promotion is human-driven.

Domain-separated configurations

Cross-domain logic is a senior architect call.

If you're being told an AI tool can do all of this end to end, ask for proof. We don't claim it because it isn't true.

05The agent roster

152 specialists. One coordinated team.

GeneWorks isn't one model wearing different hats. It's a roster of specialist agents — each scoped to a discipline, each with their own tooling, each with the prompts and patterns that produce reliable output for their artifact type. Gene is the orchestrator that routes work between them.

Group 1

Strategy & Design

5 agents
Functional ConsultantSolution ArchitectPlatform ArchitectProcess DesignerIntegration Lead
Group 2

Build

11 agents
ITSM SpecialistITOM SpecialistCSM SpecialistHRSD SpecialistSecOps SpecialistGRC AnalystServiceNow DeveloperScoped App EngineerFlow Designer SpecialistIntegrationHub EngineerService Portal Developer
Group 3

Quality

5 agents
ATF TesterSelenium TesterQA EngineerUAT AnalystPerformance Engineer
Group 4

Documentation & Compliance

5 agents
GenewriterGeneDocsCompliance AdvisorRelease ManagerChange Coordinator
Group 5

Operations & Knowledge

6 agents + Gene (orchestrator)
CMDB StewardKnowledge CuratorEvidence CollectorSelf-Healing AgentDependency MapperWorkspace CoordinatorGene · Orchestrator
06Module coverage

Where the loop runs today.

The continuous loop is live across ITSM and ITOM with full agent support. CSM, HRSD, FinOps, and SecOps follow on the published roadmap. Below is what's available now — and what each workspace handles end-to-end.

ITSM Workspace · Live

Incident, Problem, Change, SLM, Knowledge

End-to-end ITSM delivery in the continuous loop — catalog item creation, business rule modifications, SLA reconfiguration, knowledge article authoring, change management workflow updates. Full ATF coverage on every build, every story, every release.

IncidentProblemChangeSLMKnowledgeService Catalog
ITOM Workspace · Live

Discovery, Service Mapping, Event Management, AIOps

The ITOM agent handles event rule generation, alert correlation logic, threshold tuning, and CMDB CI class extensions. Service Mapping pattern configuration sits in the medium-fit tier; the underlying topology and dependency authoring stays human-owned.

Event MgmtAlert correlationCMDB CI extensionAIOps thresholds
CSM Workspace · Q3 2026

Customer Service Management

Case management, omni-channel routing, account hierarchy, customer portal builds. Same continuous loop, same two human gates, same ATF coverage. Cross-workspace dependency awareness flags conflicts with ITSM routing automatically.

Case mgmtOmni-channelCustomer Portal
HRSD Workspace · Q3 2026

HR Service Delivery

Employee lifecycle workflows, onboarding orchestration, HR case management, employee document templates. Bidirectional dependency tracking with ITSM (provisioning), CSM (employee-as-customer flows), and downstream identity systems.

Employee onboardingHR casesDocument templates
Integrations · Live

REST, Spokes, Inbound & Outbound

The Integration agents handle Outbound REST messages, Scripted REST APIs (inbound), IntegrationHub spoke configuration, and transform maps. Authentication flow design, secret management, and gateway sizing remain human-driven.

Outbound RESTScripted REST APIIntegrationHub spokesTransform maps
Custom Apps · Live

Scoped Application Engineering

Scoped app scaffolding, table design, business rules, client scripts, integration patterns, basic Service Portal widgets, full ATF coverage. The agent team delivers the build with documentation and tests included — not bolted on after.

Scoped appsTable designServer / client logicATF coverage

This is the future. The future is Now.