Workstations
Choose and configure you-agent-factory workstations across two independent axes: WorkstationType (runtime implementation) and WorkstationKind behavior (scheduling), with a type-versus-behavior compatibility matrix and shared-field summary.
A workstation binds a named worker, owns step routing and limits, and combines a runtime type with a scheduling behavior. Pick both axes independently — especially keep type POLLER_RUN distinct from behavior POLLER — then open the matching variant page.
How To Use
Use the type and behavior selection tables to open the matching variant pages. WorkstationType (field type) selects the runtime implementation family such as INFERENCE_RUN or AGENT_RUN. WorkstationKind (field behavior) selects scheduling such as STANDARD, REPEATER, CRON, or POLLER. The two axes combine freely: every published type is compatible with every published behavior. POLLER_RUN is a runtime type; POLLER is a scheduling behavior — they are not the same discriminator. Author shared fields such as name, worker, inputs, type, behavior, outputs, onFailure, and limits before variant-specific configuration.Select A Runtime Type
Use the comparison below to open the matching WorkstationType page. Choose the runtime implementation that matches how the step should execute, then pick a behavior on the scheduling axis.| WorkstationType | Use when | Variant page |
|---|---|---|
| INFERENCE_RUN | Dispatch one-shot inference harness work through an inference worker | Inference run |
| AGENT_RUN | Dispatch agent-loop work through an agent worker with optional OpenCode overrides | Agent run |
| SCRIPT_RUN | Dispatch command-backed work through a script worker | Script run |
| POLLER_RUN | Execute poller-worker runtime work — distinct from scheduling behavior POLLER | Poller run |
| MODEL_WORKSTATION | Model-oriented prompts, outcome formats, output schemas, and stop words | Model workstation |
| MODEL_INVOKE | Provider-neutral operation bindings and model operations | Model invoke |
| LOGICAL_MOVE | Non-worker logical routing with optional guarded loop breakers | Logical move |
| CLASSIFIER_WORKSTATION | Classification routes with classifier-specific output restrictions | Classifier |
Select A Scheduling Behavior
Use the comparison below to open the matching WorkstationKind behavior page. Behavior controls when and how the step becomes ready; it does not replace the runtime type.| WorkstationKind | Use when | Variant page |
|---|---|---|
| STANDARD | Ordinary readiness, dispatch, concurrency, and routing | Standard |
| REPEATER | Change-triggered looping and rejection reloop semantics | Repeater |
| CRON | Cron configuration, time work, guards, and input readiness | Cron |
| POLLER | Long-lived poller scheduling — distinct from runtime type POLLER_RUN | Poller |
Do not collapse POLLER_RUN (runtime type) with POLLER (scheduling behavior). Configure both axes when a poller step needs poller-worker execution and poller scheduling.
Type Versus Behavior Compatibility
Type and behavior are independent. Every cell below is a valid pairing: choose any WorkstationType with any WorkstationKind. The matrix is selection guidance, not a second schema contract.| WorkstationType | STANDARD | REPEATER | CRON | POLLER |
|---|---|---|---|---|
| INFERENCE_RUN | Compatible | Compatible | Compatible | Compatible |
| AGENT_RUN | Compatible | Compatible | Compatible | Compatible |
| SCRIPT_RUN | Compatible | Compatible | Compatible | Compatible |
| POLLER_RUN | Compatible | Compatible | Compatible | Compatible |
| MODEL_WORKSTATION | Compatible | Compatible | Compatible | Compatible |
| MODEL_INVOKE | Compatible | Compatible | Compatible | Compatible |
| LOGICAL_MOVE | Compatible | Compatible | Compatible | Compatible |
| CLASSIFIER_WORKSTATION | Compatible | Compatible | Compatible | Compatible |
Workstation Schema
The embed shows the live Factory Workstation base definition from the installed package. Open the full Factory schema reference for exhaustive field lookup rather than copying contract text into this index.Workstation
objectA processing step in the factory graph. Workstations consume authored work states, run a worker or logical move, and emit the next work states.
- additionalProperties
false (closed)
Fields
Scheduling behavior for this workstation, such as STANDARD, REPEATER, or CRON execution.
- body
bodyOptionalstringInline workstation instructions or script body when authored directly in factory config.
Explicit label-to-destination routing used only by CLASSIFIER_WORKSTATION definitions. Each route must declare a unique non-empty label and one or more outputs.
- copyReferencedScripts
copyReferencedScriptsOptionalbooleanCopy supported referenced script files into the expanded workstation layout when config expand runs.
Cron trigger configuration for workstations whose behavior is CRON.
Optional localized customer-facing explanation of this workstation.
Environment variables added to the workstation execution context.
Guarded loop breakers should use `VISIT_COUNT` guards here with a `LOGICAL_MOVE` workstation instead of top-level exhaustion rules.
- id
idOptionalstringOptional durable public identifier for this workstation. Graph and layout references should use this id instead of the mutable name.
Work states this workstation can consume before it dispatches.
Retry and execution ceilings applied to this workstation.
- name
nameRequiredstringCustomer-authored workstation name used by guards, diagnostics, and authored references.
Optional destination emitted when the workstation makes partial progress and should continue iterating. Classifier workstations must not declare onContinue.
Optional destination emitted when the workstation fails permanently.
Optional destination emitted when the worker rejects the current work without a hard failure. Classifier workstations must not declare onRejection.
Uppercase provider-agnostic operation requested by `MODEL_INVOKE` workstations, such as `TTS`.
Optional workstation-authored slot bindings that resolve operation inputs from runtime content or static config content.
Optional worker-output parsing mode for model workstations. When set to `decision-envelope`, agent output is parsed as a reviewer/checker JSON envelope that maps directly onto WorkResult outcome, feedback, output, and optional recorded output work instead of stop-token routing.
- outputSchema
outputSchemaOptionalstringJSON schema string used to validate or parse structured model output when configured.
Work states emitted after a non-classifier workstation succeeds. Classifier workstations must use classificationRoutes instead of normal success outputs.
- promptFile
promptFileOptionalstringPath to a prompt template file loaded for model-oriented workstation execution.
Resource capacity this workstation consumes while one dispatch is in flight.
Optional workstation-specific runner override. When omitted, dispatch falls back to the factory runner, then worker modelProvider compatibility when no explicit runner is configured, then the default codex runner.
- stopWords
stopWordsOptionalarrayStop words authored on the topology entry for model-oriented dispatches.
Runtime workstation implementation type, equivalent to the workstation AGENTS.md frontmatter type.
Optional policy for whether downstream work uses the workstation output payload or preserves the consumed input payload.
- worker
workerRequiredstringName of a worker declared in the workers list.
- workingDirectory
workingDirectoryOptionalstringGo template resolved from token tags at dispatch time.
- worktree
worktreeOptionalstringGo template resolved and passed as the worktree path to CLI dispatchers.