Menü
Run the full static analyzer on a BPMN file and report deploy-readiness. Usage: /bpmnkit:review [file.bpmn]
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
Design and add a Camunda AI Agent Sub-process to a new or existing BPMN process — model, prompts, tools, and a mocked test scenario. Usage: /bpmnkit:agent [file] <description>
Wire an existing BPMN step to an external service via a Camunda connector template. Usage: /bpmnkit:connect <file> <step-id> <service>
Gate a BPMN process on deploy-readiness, then deploy it to local Reebe or Camunda 8. Usage: /bpmnkit:deploy [file.bpmn] [--local|--camunda]
Extend an existing BPMN process from a natural-language change request — lifts it to a plan, applies a targeted delta, and merges. Usage: /bpmnkit:extend <file.bpmn> <change request>
Implement a BPMN process end-to-end from a natural language description — writes a ProcessPlan, compiles it deterministically, tests it, and offers to deploy. Usage: /bpmnkit:implement <description>
Run scenario tests on a BPMN process file and report path coverage. Usage: /bpmnkit:test [file.bpmn]
| name | review |
| description | Run the full static analyzer on a BPMN file and report deploy-readiness. Usage: /bpmnkit:review [file.bpmn] |
Review the BPMN file for issues: $ARGUMENTS
If $ARGUMENTS names a .bpmn file, use it. Otherwise run ls *.bpmn in the current directory: if exactly one exists, use it; if multiple, list them and ask.
casen lint lint <file>.bpmn --profile deploy --format json
casen lint lint <file>.bpmn --format json
(lint's flag is --format, not --output — don't confuse it with synth's --output <path>, which writes to a file and is a different flag on a different command.)
The first is the deploy-readiness gate (errors only, across every category: flow, naming, feel, feel-syntax, deploy, agentic, connector, pattern, data-flow, task-reuse, extract). The second is the full report including warnings/info.
**Errors** (must fix before deploy)
- [element-id] [category] message. Fix: suggestion, if present.
**Warnings** (should fix)
- ...
**Info** (consider)
- ...
Ignore data-flow info/warning findings against an aiAgent step's own provider/model/prompt bindings — see references/agentic.md's "Known noisy finding" note; they're a known analyzer limitation, not a real issue.
If there are any findings at all, offer to run:
casen lint lint <file>.bpmn --fix
This applies every auto-fixable finding and writes the result back (the CLI reports how many were fixed, or "No auto-fixable issues found."). Re-run step 2 afterward to confirm.
End with one line: "Deploy-ready: yes" (zero errors from the deploy profile) or "Deploy-ready: no — N error(s)".