Automate Zeebe BPMN Process

VerifiedSafe

This skill generates hexagonal-architecture glue-code for automating a Zeebe BPMN process, creating job workers, a process out-adapter, inbound/outbound port interfaces, and application service stubs. It accepts a ProcessApi file, BPMN model, or plain description, and skips existing files. Use it when you need to scaffold workers and adapters for a BPMN process.

Sby Skills Guide Bot
DevelopmentIntermediate
6/2/2026
Claude Code
#zeebe#bpmn#hexagonal-architecture#code-generation

Recommended for

Automated summary

Generates complete hexagonal glue-code for automating a Zeebe BPMN process, including workers, adapters, ports, and services.

Strengths

  • Produces a consistent hexagonal architecture with clear layer separation.
  • Uses ProcessApi constants to avoid raw string literals.
  • Offers optional BPMN validation via review-process sub-agent before generation.

Limitations

  • Requires an existing ProcessApi file or user to generate one first.
  • Does not implement business logic, only stubs for application services.
  • Execution may pause if user chooses to run the review sub-agent.
When to use it

When you need to quickly scaffold the infrastructure code for automating a Zeebe BPMN process with hexagonal architecture.

When not to use it

When the BPMN process is not yet finalized or you prefer to manually write each layer.

Security analysis

Safe
Quality score90/100

The skill generates boilerplate code based on a ProcessApi file using read/write tools. It does not execute any system commands, exfiltrate data, or engage in network communication. The skill workflow is deterministic and file-scoped, posing no direct security risk.

No concerns found

How this analysis is made

Examples

Automate order process from BPMN
Automate the order process using the BPMN model in resources/order-process.bpmn
Generate glue-code from ProcessApi
Generate glue-code for the ProcessApi file at adapter/process/OrderProcessApi.kt
Scaffold workers and adapters for a process
Scaffold workers and adapters for the customer-onboarding process

name: automate-process argument-hint: "[<path-to-ProcessApi-or-BPMN-file>]" allowed-tools: Read, Write, Glob description: Generate full hexagonal glue-code for a Zeebe BPMN process across all five layers — inbound workers, inbound/outbound port interfaces, application services, and a process out-adapter. Use when the user asks to "automate a process", "generate glue-code for a BPMN process", or "scaffold workers and adapters". Accepts a ProcessApi file, a BPMN model, or a plain description; skips files that already exist; optionally runs the review-process audit first.

Skill: automate-process

Generate full hexagonal-architecture glue-code for automating a Zeebe BPMN process: job workers, process out-adapter, inbound/outbound port interfaces, and application service stubs.

What This Skill Creates

// TODO: use a list instead of a layer like

  • Inbound Workers: stored in adapter/inbound/zeebe with one @JobWorker per serviceTask

| Layer | Location | What | |----------------------|------------------------------|------------------------------------------| | Inbound workers | adapter/inbound/zeebe/ | One @JobWorker class per service task | | Outbound adapter | adapter/outbound/zeebe/ | One process out-adapter class | | Inbound ports | application/port/inbound/ | One use-case interface per worker | | Outbound port | application/port/outbound/ | One process port interface | | Application services | application/service/ | One service class per use-case interface |

The adapter and worker files reference these interfaces so the hexagonal wiring is complete from the start.

IMPORTANT

All string constants must come from the ProcessApi — never use raw string literals.

  • Worker @JobWorker(type = ...) → ProcessApi.TaskTypes.CONSTANT
  • Adapter startProcess(processId = ...) → ProcessApi.PROCESS_ID
  • Adapter sendMessage(messageName = ...) → ProcessApi.Messages.CONSTANT
  • Variable keys → ProcessApi.Variables.CONSTANT where defined

Instructions

Preparation – Validate the BPMN model first

Before generating any code, ask the user whether he has already performed the review-process subagent - and ask him whether you should do it. This is very useful, to only generate code for clean processes that follow the styleguide.

If the user agrees to running the subagent, then start it, but terminate the execution of this skill. Tell the user to wait until the review-process agent is done.

After that he must fully restart the automate-process once again If he disagrees to perform the review agent, continue with step 1.

Step 1 – Resolve the input source

Determine where the process constants come from based on $ARGUMENTS:

  • ProcessApi file (.kt): read it directly. Extract: package name, object name, PROCESS_ID, all TaskTypes.*, Messages.*, Signals.* (if present), and Variables.* constants.
  • BPMN file (.bpmn): search the same service module for a *ProcessApi.kt file (Glob **/adapter/process/*ProcessApi.kt). If found, read it as above. If not found, ask the user whether to continue without type-safe constants.
  • No argument / plain description: search the whole codebase for *ProcessApi.kt files (Glob **/adapter/process/*ProcessApi.kt). List them and ask the user which one to use.

If you can't find a ProcessApi, pause the execution of the skill, and ask the user to generate one first.

Step 2 – Determine packages and source root

Derive the base package from the ProcessApi package. Example:

  • ProcessApi: io.miragon.example.adapter.process
  • Base: io.miragon.example
  • Inbound workers: <base>.adapter.inbound.zeebe
  • Outbound adapter: <base>.adapter.outbound.zeebe
  • Inbound ports: <base>.application.port.inbound
  • Outbound port: <base>.application.port.outbound
  • Services: <base>.application.service

Step 2.5 – Discover existing domain types

Before generating any port or adapter code, scan the domain package (<base>.domain) for existing value objects.

For each Variables.* constant in the ProcessApi:

  • Derive a candidate domain type name by converting the constant to PascalCase (e.g. NEWSLETTER_ID → NewsletterId, SUBSCRIPTION_ID → SubscriptionId)
  • If a variable name makes no sense as a domain type name, ask the developer for feedback
  • Check if a file with that name exists under **/domain/<TypeName>.kt
  • If found, read it and note the class name and its primary constructor parameter type (typically UUID or String)

If no suitable domain type exists for a variable, create one (a Kotlin data class wrapping UUID or String, with a secondary String constructor when the primary wraps UUID) before generating ports or adapters.

Step 3 – Generate inbound port interfaces (use cases)

For each TaskTypes.* constant, generate a use-case interface. Name convention: derive a meaningful interface name from the task type constant (e.g. NEWSLETTER_ABORT_REGISTRATION → AbortSubscriptionUseCase).

Generated structure:

  • Interface with one method matching the worker's action
  • Method parameters must use domain types discovered in Step 2.5 — never raw String or UUID

Skip if file already exists.

Step 4 – Generate outbound port interface

Generate one outbound port interface for the process. Name convention: derive from the process name (e.g. NewsletterSubscription → NewsletterSubscriptionProcess).

Generated structure:

  • startProcess method (returns Long) — or a message-send method if the process uses a message start event
  • One method per Messages.* constant
  • All method parameters use domain types discovered in Step 2.5

Skip if file already exists.

Step 5 – Generate application services

For each inbound use-case interface, generate an application service implementing it. Name convention: append Service to the use-case name (e.g. AbortSubscriptionUseCase → AbortSubscriptionService).

Generated structure:

  • @Service class implementing the use-case interface
  • override fun with domain type parameters (matching the use-case interface) and TODO("Add implementation") body

Skip if file already exists.

Step 6 – Generate workers (with use-case injection)

Same as /create-worker Step 3, but inject the generated use-case interface as the constructor parameter instead of a placeholder. Apply the same @Variable / @VariableAsType guidance from that step.

Worker handle method parameters must be primitive types (UUID, String) annotated with @Variable; convert them to domain types inside the body before calling the use case (e.g. useCase.handle(NewsletterId(newsletterId))).

Step 7 – Generate process out-adapter (implementing outbound port)

Same as /create-process-adapter Step 3, but add : ProcessNameProcess to the class declaration so it implements the outbound port interface. Mark each method with override.

Adapter methods must accept domain types as parameters. Inside the method body, extract the primitive value (e.g. id.value.toString()) before passing to engineApi.startProcess / engineApi.sendMessage.

Step 8 – Report

List every file created or skipped. Remind the developer to:

  1. Run /test-worker for each worker
  2. Run /test-process-adapter for the adapter
  3. Run /test-process to generate process integration tests

Offer to run these test-generation commands right away.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/automate-zeebe-bpmn-process-mm8o5324/download' --output '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md.part' && mv '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md.part' '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Tools declared by the author (check availability in your agent): Read, Write, Glob

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md' && test -s '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/automate-process' && cp -R '.skill-downloads/automate-zeebe-bpmn-process-mm8o5324' '.claude/skills/automate-process'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/automate-zeebe-bpmn-process-mm8o5324'

openskills read 'automate-zeebe-bpmn-process-mm8o5324'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/automate-zeebe-bpmn-process-mm8o5324/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills