Scenario Workflow Authoring
Overview
A workflow has two representations: editor_info (the editable node graph: nodes, edges,
inputKeys) and flow (the compiled runnable form). Authoring through MCP means writing the whole
editor_info document: there are no per-node editing tools; every change is a read, modify, write
of the full graph through workflow_create or workflow_update. Never hand-write flow:
workflow_publish compiles editor_info into it and flips status to ready. Editing a ready
workflow's editor_info leaves the stale flow running until you publish again.
Read references/editor-info.md before writing any graph: it holds the
node type vocabulary, the node choice doctrine (when an llm node is legitimate), the edge
direction rule, per-node data contracts, and a validated minimal example. Create, update, publish,
copy and delete live in the tool catalog (scenario_tools_search plus the matching executor, see
the scenario skill). workflows_list, workflow_get and workflow_run are direct tools: scope
and dry_run go in their top-level arguments, never an executor wrapper. Running and pricing: the
scenario-workflows skill. If a sibling skill named here is missing from your available skills,
load it with skill_get (Scenario MCP) or ask the user to install it
(https://github.com/scenario-labs/skills#install); unattended, proceed from tool schemas and flag
the gap.
Quick reference
| Step |
Call |
Notes |
| 1. Study a graph |
workflow_get on a working workflow |
Copy the shape, never ids |
| 2. Model contract |
model_schema_get |
Handle names and required inputs |
| 3. Author |
editor_info + inputs_definition |
Per the reference file |
| 4. Create |
workflow_create |
Non-atomic, see below |
| 5. Publish |
workflow_publish |
Compiles flow; one output pin required |
| 6. Validate |
workflow_run with dry_run=true |
Prices and runs the real validator |
workflow_create is two calls under the hood: a failed create may still have created a draft whose
id is in the error. Recover with workflow_update on that id; re-creating duplicates. Seed step 1
with workflow_get: it returns the full graph of any workflow whose id you have, public ones
included (an id or app URL the user supplies, or your own team's from workflows_list). To find a
public template, use search with target="workflows", public=true, a keyword query,
limit=3, and your scope; for example, query="image" with raw filter: 'status = "ready"'. Read
ids from workflows, then fetch the chosen graph with workflow_get; search hits are summaries,
not graph documents. Workflow search supports keyword text and filters only, so omit image and
semantic options. Use workflows_list for browsing your saved workflows.
scripts/fetch_workflow_examples.py bulk-exports trimmed
featured-workflow graphs for maintainers (setup in its header).
Worked example: a text-to-image app
recommend with capability: "txt2img" and the user's brief as prompt, following the
scenario skill's next_step discipline, then model_schema_get: its input names become the
model node's handle names, and its required flag marks what must be wired. Use search instead
when the user names a model.
- Author
editor_info: text1 with data.isInput: true, model1 with type: "model",
data.modelId and data.isOutput: true, one edge from model1's input to text1's output
(edges name the downstream node as source, see the reference), inputKeys: ["text1"].
workflow_create with name, editor_info, and inputs_definition naming text1 as a string
input. The published input key is the node id, which is why run inputs have names like text1.
workflow_publish, then workflow_run with dry_run=true to validate and price. Fix the graph
and re-publish if validation fails.
One output per item needs a loop
A multiple-asset input wired straight into a generative model's reference field is one generation
conditioned on every item at once: ten product shots in come back as one image blending them, not
ten edits. Field names and cardinality are per model, so read them off model_schema_get
(array: true marks a list), and read the field's description as well: a few utility tools take a
list and return one output per item, which needs no loop. When each item needs its own result, put
the model inside a forEach over the list and read the outputs from the forEachEnd (wiring in the
reference's forEach section); keep the direct wiring only when the items are meant as joint
references for a single output. Inside the loop the current item feeds an array field as a one-item
list. forEach runs one billed model job per item, but the dry run prices one iteration whatever
the list length (observed: one quote for one or two items, twice that billed for two), so multiply
the quote by the item count before quoting the user.
Migrating a graph from another node tool
A pipeline exported by Weavy, ComfyUI, or another node editor does not import: only Scenario's own
export round-trips. It is translated node by node, then created, published, and dry-run as above.
The mapping table, member resolution, and the report the user gets are in
references/foreign-graph-import.md; read it before touching
such an export, since its first rule is to reduce the file to a table locally rather than paste it
into the conversation.
Common mistakes
- Writing UI palette names as node types: persisted types are the camelCase vocabulary in the
reference, and every generator is
type: "model".
- Wiring edges producer to consumer: persisted edges point the other way.
- Expecting an
editor_info update to change a live app without re-publishing.
- Retrying a failed
workflow_create with a second create instead of workflow_update on the id
from the error.
- Wiring a batch of assets into one model input and expecting one output per item: that is one
generation over all of them; loop with
forEach.
- Publishing with no output pin, or with
data.isOutput on a text or asset node: publish fails with
compiled to an empty flow. The output pin goes on the model, llm, or forEachEnd node that
produces the deliverable. Input pins are optional: a graph with its prompt built in publishes and
runs with no inputs.
- Gating a text node in front of a builder or model: a branch skips only the node wired to its
handle, so the consumer stays pending and the job never completes. Gate the node that does the
work, or use a CEL ternary for conditional prompt text, per the reference's
ifElse section.
- Double-quoted CEL literals: they evaluate but corrupt the canvas editor, single quotes only.
- Sending
workflow_id to workflow_copy: get, update, publish, run and delete take workflow_id,
but copy takes source_workflow_id; the copy inherits everything verbatim and needs its own
publish.
1---2name: scenario-workflow-authoring3description: "Use when a task involves creating or editing a Scenario workflow graph through MCP: building an app from a brief, adding or rewiring nodes (models, prompts, approval gates, loops), authoring editor_info, publishing, unpublishing or renaming, importing an exported workflow JSON, migrating a graph built in Weavy, ComfyUI, or another node tool, copying a workflow, or turning a prompt chain into an app. Running or pricing a workflow is scenario-workflows. Keywords: node graph, editor_info, CEL."4license: MIT5metadata:6 source: https://skillmd.com/skills/scenario-labs/scenario-workflow-authoring7---89# Scenario Workflow Authoring1011## Overview1213A workflow has two representations: `editor_info` (the editable node graph: `nodes`, `edges`,14`inputKeys`) and `flow` (the compiled runnable form). Authoring through MCP means writing the whole15`editor_info` document: there are no per-node editing tools; every change is a read, modify, write16of the full graph through `workflow_create` or `workflow_update`. Never hand-write `flow`:17`workflow_publish` compiles `editor_info` into it and flips status to `ready`. Editing a ready18workflow's `editor_info` leaves the stale `flow` running until you publish again.1920Read [references/editor-info.md](references/editor-info.md) before writing any graph: it holds the21node type vocabulary, the node choice doctrine (when an `llm` node is legitimate), the edge22direction rule, per-node data contracts, and a validated minimal example. Create, update, publish,23copy and delete live in the tool catalog (`scenario_tools_search` plus the matching executor, see24the `scenario` skill). `workflows_list`, `workflow_get` and `workflow_run` are direct tools: scope25and `dry_run` go in their top-level arguments, never an executor wrapper. Running and pricing: the26`scenario-workflows` skill. If a sibling skill named here is missing from your available skills,27load it with `skill_get` (Scenario MCP) or ask the user to install it28(https://github.com/scenario-labs/skills#install); unattended, proceed from tool schemas and flag29the gap.3031## Quick reference3233| Step | Call | Notes |34| ----------------- | ------------------------------------ | ---------------------------------------- |35| 1. Study a graph | `workflow_get` on a working workflow | Copy the shape, never ids |36| 2. Model contract | `model_schema_get` | Handle names and required inputs |37| 3. Author | `editor_info` + `inputs_definition` | Per the reference file |38| 4. Create | `workflow_create` | Non-atomic, see below |39| 5. Publish | `workflow_publish` | Compiles `flow`; one output pin required |40| 6. Validate | `workflow_run` with `dry_run=true` | Prices and runs the real validator |4142`workflow_create` is two calls under the hood: a failed create may still have created a draft whose43id is in the error. Recover with `workflow_update` on that id; re-creating duplicates. Seed step 144with `workflow_get`: it returns the full graph of any workflow whose id you have, public ones45included (an id or app URL the user supplies, or your own team's from `workflows_list`). To find a46public template, use `search` with `target="workflows"`, `public=true`, a keyword `query`,47`limit=3`, and your scope; for example, `query="image"` with raw `filter: 'status = "ready"'`. Read48ids from `workflows`, then fetch the chosen graph with `workflow_get`; search hits are summaries,49not graph documents. Workflow search supports keyword text and filters only, so omit image and50semantic options. Use `workflows_list` for browsing your saved workflows.51[scripts/fetch_workflow_examples.py](scripts/fetch_workflow_examples.py) bulk-exports trimmed52featured-workflow graphs for maintainers (setup in its header).5354## Worked example: a text-to-image app55561. `recommend` with `capability: "txt2img"` and the user's brief as `prompt`, following the57 `scenario` skill's `next_step` discipline, then `model_schema_get`: its input names become the58 model node's handle names, and its `required` flag marks what must be wired. Use `search` instead59 when the user names a model.602. Author `editor_info`: `text1` with `data.isInput: true`, `model1` with `type: "model"`,61 `data.modelId` and `data.isOutput: true`, one edge from `model1`'s input to `text1`'s output62 (edges name the downstream node as `source`, see the reference), `inputKeys: ["text1"]`.633. `workflow_create` with `name`, `editor_info`, and `inputs_definition` naming `text1` as a string64 input. The published input key is the node id, which is why run inputs have names like `text1`.654. `workflow_publish`, then `workflow_run` with `dry_run=true` to validate and price. Fix the graph66 and re-publish if validation fails.6768## One output per item needs a loop6970A multiple-asset input wired straight into a generative model's reference field is one generation71conditioned on every item at once: ten product shots in come back as one image blending them, not72ten edits. Field names and cardinality are per model, so read them off `model_schema_get`73(`array: true` marks a list), and read the field's description as well: a few utility tools take a74list and return one output per item, which needs no loop. When each item needs its own result, put75the model inside a `forEach` over the list and read the outputs from the `forEachEnd` (wiring in the76reference's forEach section); keep the direct wiring only when the items are meant as joint77references for a single output. Inside the loop the current item feeds an array field as a one-item78list. `forEach` runs one billed model job per item, but the dry run prices one iteration whatever79the list length (observed: one quote for one or two items, twice that billed for two), so multiply80the quote by the item count before quoting the user.8182## Migrating a graph from another node tool8384A pipeline exported by Weavy, ComfyUI, or another node editor does not import: only Scenario's own85export round-trips. It is translated node by node, then created, published, and dry-run as above.86The mapping table, member resolution, and the report the user gets are in87[references/foreign-graph-import.md](references/foreign-graph-import.md); read it before touching88such an export, since its first rule is to reduce the file to a table locally rather than paste it89into the conversation.9091## Common mistakes9293- Writing UI palette names as node types: persisted types are the camelCase vocabulary in the94 reference, and every generator is `type: "model"`.95- Wiring edges producer to consumer: persisted edges point the other way.96- Expecting an `editor_info` update to change a live app without re-publishing.97- Retrying a failed `workflow_create` with a second create instead of `workflow_update` on the id98 from the error.99- Wiring a batch of assets into one model input and expecting one output per item: that is one100 generation over all of them; loop with `forEach`.101- Publishing with no output pin, or with `data.isOutput` on a text or asset node: publish fails with102 `compiled to an empty flow`. The output pin goes on the `model`, `llm`, or `forEachEnd` node that103 produces the deliverable. Input pins are optional: a graph with its prompt built in publishes and104 runs with no inputs.105- Gating a text node in front of a builder or model: a branch skips only the node wired to its106 handle, so the consumer stays pending and the job never completes. Gate the node that does the107 work, or use a CEL ternary for conditional prompt text, per the reference's `ifElse` section.108- Double-quoted CEL literals: they evaluate but corrupt the canvas editor, single quotes only.109- Sending `workflow_id` to `workflow_copy`: get, update, publish, run and delete take `workflow_id`,110 but copy takes `source_workflow_id`; the copy inherits everything verbatim and needs its own111 publish.