Pizza Developer

Update workflow

update_workflow

Destructive

Edit one workflow. Pass nodes and edges together to replace its graph, and name or description to change its details; omitted parts stay as they are. Nodes are Start, optional Idea, Batch, optional Delivery, optional Workflow (the step that improves the Workflow itself after a run), and Finish. Start can connect to an existing Batch without Idea, and Batch can connect to Workflow or Finish without Delivery. Start can also connect to Delivery, for a workflow that only creates or adds to a Delivery: Start to Delivery to Finish, or Start to Delivery to Workflow to Finish, with no Batch node. Each edge.data must contain instructionDocumentIds (existing project document IDs) and contextKinds (features and/or dictionary; each adds only a short outline, top-level feature names or dictionary words, that the agent expands with the read tools). edge.data may also carry rulesetIds, an ordered list of the project's ruleset IDs attached to that connection; leaving it out attaches none. Select documents on connections; do not create Instructions, Dictionary, or Features nodes or put text on lines. Updates made during a run's self-improvement stage are described in the report that closes that step. Open runs keep their frozen revision. Graph writes are refused while a vibe coder has the workflow open for editing in the browser, and archived workflows must be restored first.

Input Schema

TableJSON Schema

Name	Required	Description
`name`	No	New Workflow name.
`edges`	No	Replacement connections; pass together with nodes.
`nodes`	No	Replacement stages; pass together with edges.
`projectId`	Yes	Id of the Project.
`workflowId`	Yes	Id of the Workflow.
`description`	No	New description of the requests it handles.

Schema Changelog

Changes observed during successful MCP inspections.

Changed10 schema fields changedOct 9, 2026

addedInput schema / properties / edges / items / additionalProperties
```
Added value: +{}
```

addedInput schema / properties / edges / items / properties

Added value: +{
+  "data": {
+    "additionalProperties": {},
+    "properties": {
+      "contextKinds": {
+        "description": "Outlines added on this connection.",
+        "items": {
+          "enum": [
+            "features",
+            "dictionary"
+          ],
+          "type": "string"
+        },
+        "type": "array"
+      },
+      "instructionDocumentIds": {
+        "description": "Ids of the Project's Instructions sent on this connection.",
+        "items": {
+          "minLength": 1,
+          "type": "string"
+        },
+        "type": "array"
+      },
+      "rulesetIds": {
+        "description": "Ordered ids of the Rulesets checked on this connection.",
+        "items": {
+          "minLength": 1,
+          "type": "string"
+        },
+        "type": "array"
+      }
+    },
+    "required": [
+      "instructionDocumentIds",
+      "contextKinds"
+    ],
+    "type": "object"
+  },
+  "id": {
+    "description": "Connection id, unique in the Workflow.",
+    "minLength": 1,
+    "type": "string"
+  },
+  "source": {
+    "description": "Id of the node the connection leaves.",
+    "minLength": 1,
+    "type": "string"
+  },
+  "target": {
+    "description": "Id of the node the connection enters.",
+    "minLength": 1,
+    "type": "string"
+  }
+}

addedInput schema / properties / edges / items / required

Added value: +[
+  "id",
+  "source",
+  "target",
+  "data"
+]

addedInput schema / properties / edges / items / type
```
Added value: +"object"
```
addedInput schema / properties / edges / maxItems
```
Added value: +8
```
addedInput schema / properties / nodes / items / additionalProperties
```
Added value: +{}
```

addedInput schema / properties / nodes / items / properties

Added value: +{
+  "data": {
+    "additionalProperties": {},
+    "properties": {
+      "resourceKind": {
+        "description": "Stage this node stands for.",
+        "enum": [
+          "start",
+          "idea",
+          "batch",
+          "delivery",
+          "workflow",
+          "finish"
+        ],
+        "type": "string"
+      }
+    },
+    "required": [
+      "resourceKind"
+    ],
+    "type": "object"
+  },
+  "id": {
+    "description": "Node id, unique in the Workflow.",
+    "minLength": 1,
+    "type": "string"
+  },
+  "position": {
+    "description": "Position on the canvas.",
+    "properties": {
+      "x": {
+        "type": "number"
+      },
+      "y": {
+        "type": "number"
+      }
+    },
+    "required": [
+      "x",
+      "y"
+    ],
+    "type": "object"
+  }
+}

addedInput schema / properties / nodes / items / required
```
Added value: +[
+  "id",
+  "position",
+  "data"
+]
```
addedInput schema / properties / nodes / items / type
```
Added value: +"object"
```
addedInput schema / properties / nodes / maxItems
```
Added value: +6
```

First observedOct 3, 2026

TDQS

A4.7/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare destructiveHint=true; the description adds substantial context beyond that: partial-update semantics (omitted parts stay as they are), the frozen-revision behavior of open runs, self-improvement updates being reported in the closing report, the browser-editing write lock, and the archived-workflow precondition. This is exactly the extra behavioral context annotations cannot convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The lead sentence is front-loaded and every subsequent sentence carries real information (valid connection patterns, edge requirements, preconditions), so it is not padded. It is a single dense block, though, and could be broken into clearer chunks for readability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex graph-mutating tool with no output schema, the description covers mutation semantics, valid graph shapes, edge data requirements, write-lock behavior, archived-state precondition, and run-revision behavior. An agent has what it needs to invoke this correctly without opening other tools.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although schema coverage is 100%, the description adds semantics the schema does not: the meaning of each node kind, which node-to-node connections are valid (e.g. Start to Batch without Idea, Start to Delivery with no Batch), that edge.data must carry instructionDocumentIds and contextKinds, and that rulesetIds is an ordered optional list. This materially shapes how the agent fills nodes/edges.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb+resource ("Edit one workflow") and immediately distinguishes this from creation/deletion siblings by describing the graph-replacement and detail-edit modes. An agent can tell what the tool does and how it differs from create_workflow/delete_workflow without reading the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete operating rules: pass nodes and edges together to replace the graph, omit parts to leave them unchanged, and explicit preconditions (graph writes refused while a vibe coder has it open; archived workflows must be restored first). It also states what not to do (don't create Instructions/Dictionary/Features nodes or put text on lines). It stops short of naming specific sibling alternatives for narrower edits.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Pizza Developer tools Inspect

Try in Browser

Pay for results, not tokens.Fund a bounty. Let AI agents compete and collaborate to solve your problems.agentbounties.appAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?