Skip to main content
Glama

Create a workspace node of any type

node_create

Create a workspace node of any type

If this job may already have a playbook, call playbooks search first. One call creates any of the 11 node types with its type-specific payload: fields for a Base, body for a Doc, files for a Skill/Drive/AirApp, assetId for a File. visibility works for every type: pass private when the person asked for something only they (and people they grant) should see. Review is permission-aware, decided server-side: this merges immediately when you already have write access on the parent node, and proposes a ChangeRequest for a human otherwise. Pass requireReview to always propose instead of merging. Before you finish: if the person will come back to this node to do the same job again, pass agentPrompts so the node opens with THEIR job on it instead of the node type's generic list.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "type": "<type>",
      "slug": "<slug>",
      "name": "<name>"
    },
    "name": "node_create"
  }
}

Input Schema

NameRequiredDescriptionDefault
bodyNoDoc body in Markdown (--type doc only).
nameYesHuman-readable node name.
slugYesURL-safe identifier, lowercase letters/digits/hyphens only.
typeYesNode type to create.
filesNoSeed files for a Skill/Drive/AirApp: [{"path":"SKILL.md","content":"..."}]. Layered over the default scaffold unless mergeMode is "replace" — a path you supply REPLACES the scaffold's file at that path. For --type airapp the project MUST be runnable by `npm run dev`: give package.json a "dev" script running a plain Node server (`node server.js`, Hono or node:http). A bundler dev server (Vite/webpack/Next) and any native-binary dependency CANNOT boot in the AirApp runtime, and browser files must need no build step.
fieldsNoBase fields (--type base only): [{"slug":"title","name":"Title","type":"text"}]. A Base needs at least one field.
assetIdNoBacking Asset id (--type file only). Upload the Asset first.
messageNoExplanation for the human reviewer. Conventional-commit style, e.g. "Add Products Base for the Q3 catalog".
versionNoSkill/Drive/AirApp only. Defaults to 0.1.0.
playbookNoOptional. The playbook you are following, as `kind:nodeId[:key]` from playbooks_search (e.g. `prompt:nod_123:log-visit`). Recorded on the change request so the person can see which playbook produced it.
autoMergeNoSkip review and create immediately if you have write access. Default is permission-aware: merge when you can, otherwise propose.
mergeModeNoSkill/Drive/AirApp only. "merge" (default) layers files over the default scaffold; "replace" uses only the files given.
visibilityNoWho can see the node, for every type. "private" hides it from members who have not been granted access; "workspace" shows it to every member; "public" currently behaves as "workspace". Omit to follow the space default and parent folder (a node is never more open than its strictest parent). Also applies when this call proposes a ChangeRequest instead: it takes effect when the request merges. If the server could not apply it, the result says so in `visibilityWrite` with the command that fixes it.
descriptionNoOptional node description.
submittedByNoProducer label recorded on the change.
agentPromptsNoScenario prompts shown when someone opens this node and asks an agent for something: [{"key":"log-visit","label":"Log a customer visit","body":"Add a visit record with today's date, the contact I name, and a one-line summary.","intent":"change"}]. Write what the PERSON wants in their own words ("Log a customer visit"), not the operation ("Create a record in Visits") — the node type already covers the operations. These are added after the node type's built-in scenario prompts, so 2-5 real recurring jobs help and a generic pair only repeats the defaults: omit this when you cannot name one. Write only the task: Busabase opens every prompt with a line naming the node and space, so do not write `{target}`. Stored in a second call after the node exists, so they are skipped (and reported) when this call proposes a ChangeRequest instead of creating the node; add them afterwards with `nodes set-agent-prompts`.
parentNodeIdNoParent folder node id. Omit to create at the space root.
requireReviewNoAlways propose a pending ChangeRequest, even with write access.
targetSpaceIdNoBusabase space id. Call auth_verify first and ask the user which space to use when more than one is returned.
skipDefaultViewNoBase only. Create the Base WITHOUT its default "Grid view" (a table of every record and field). Only when you add your own views right after with view_change_request — a Base must keep at least one view, and the first one is what it opens on.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / visibility / description
      Previous value: -"Skill/Drive/AirApp only. Omit to follow the space default and parent folder, like any other node."New value: +"Who can see the node, for every type. \"private\" hides it from members who have not been granted access; \"workspace\" shows it to every member; \"public\" currently behaves as \"workspace\". Omit to follow the space default and parent folder (a node is never more open than its strictest parent). Also applies when this call proposes a ChangeRequest instead: it takes effect when the request merges. If the server could not apply it, the result says so in `visibilityWrite` with the command that fixes it."
  2. Changed1 schema field changed
    • addedInput schema / properties / skipDefaultView
      Added value: +{
      +  "description": "Base only. Create the Base WITHOUT its default \"Grid view\" (a table of every record and field). Only when you add your own views right after with view_change_request — a Base must keep at least one view, and the first one is what it opens on.",
      +  "type": "boolean"
      +}
  3. Changed1 schema field changed
    • changedInput schema / properties / agentPrompts / description
      Previous value: -"Scenario prompts shown when someone opens this node and asks an agent for something: [{\"key\":\"log-visit\",\"label\":\"Log a customer visit\",\"body\":\"{target}\\n\\nAdd a visit record with today's date, the contact I name, and a one-line summary.\",\"intent\":\"change\"}]. Write what the PERSON wants in their own words (\"Log a customer visit\"), not the operation (\"Create a record in Visits\") — the node type already covers the operations. These REPLACE the node type's default scenario prompts, so 2-5 real recurring jobs help and a generic pair is worse than none: omit this when you cannot name one. `{target}` expands to a complete sentence naming the node and space, so give it its own line. Stored in a second call after the node exists, so they are skipped (and reported) when this call proposes a ChangeRequest instead of creating the node; add them afterwards with `nodes set-agent-prompts`."New value: +"Scenario prompts shown when someone opens this node and asks an agent for something: [{\"key\":\"log-visit\",\"label\":\"Log a customer visit\",\"body\":\"Add a visit record with today's date, the contact I name, and a one-line summary.\",\"intent\":\"change\"}]. Write what the PERSON wants in their own words (\"Log a customer visit\"), not the operation (\"Create a record in Visits\") — the node type already covers the operations. These are added after the node type's built-in scenario prompts, so 2-5 real recurring jobs help and a generic pair only repeats the defaults: omit this when you cannot name one. Write only the task: Busabase opens every prompt with a line naming the node and space, so do not write `{target}`. Stored in a second call after the node exists, so they are skipped (and reported) when this call proposes a ChangeRequest instead of creating the node; add them afterwards with `nodes set-agent-prompts`."
  4. Changed1 schema field changed
    • changedInput schema / properties / visibility / description
      Previous value: -"Skill/Drive/AirApp only. Defaults to private."New value: +"Skill/Drive/AirApp only. Omit to follow the space default and parent folder, like any other node."
  5. Changed1 schema field changed
    • addedInput schema / properties / playbook
      Added value: +{
      +  "description": "Optional. The playbook you are following, as `kind:nodeId[:key]` from playbooks_search (e.g. `prompt:nod_123:log-visit`). Recorded on the change request so the person can see which playbook produced it.",
      +  "type": "string"
      +}
  6. First observed

TDQS

A4.3/5.0
Behavior5/5

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

Annotations only say this is a non-read-only, non-destructive, non-open-world write; the description goes much further, disclosing that review is permission-aware and decided server-side (merge with write access, ChangeRequest otherwise), that visibility takes effect only when a proposed request merges, and that a failed visibility write is reported in `visibilityWrite`. It also flags that agentPrompts are applied in a second call and are skipped when a ChangeRequest is proposed — behavior an agent could not otherwise predict.

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?

Front-loaded with the verb and scope, and every paragraph carries actionable guidance rather than filler. It is still a dense, long block for a single call, and the payload-key enumeration duplicates schema content that the agent will read anyway.

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

Completeness4/5

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

For a 20-parameter write tool with no output schema, the description covers the review/merge mechanics, visibility semantics, and one result field (`visibilityWrite`), which is what an agent most needs before calling. It never says what a successful creation returns (e.g., the new node id), leaving a small gap since no output schema exists to fill it.

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

Parameters3/5

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

Schema coverage is 100% and the schema descriptions are unusually thorough, so the baseline is 3. The description's per-type payload mapping (`fields`/`body`/`files`/`assetId`) largely restates enum and property descriptions; its added value is limited to framing (`visibility` works for every type) and the judgment call on when to omit agentPrompts.

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?

States a specific verb+resource ("Create a workspace node") and immediately scopes it ("One call creates any of the 11 node types"), listing the type-specific payload keys. An agent can tell this apart from node_archive, node_get_file_tree and the change-request siblings without opening a 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 when-to guidance: check playbooks search first if a playbook may exist, pass `private` when the person wants something restricted, pass requireReview to force a proposal, and pass agentPrompts when the person will return to the node. What's missing is an explicit statement of when to prefer the sibling change-request path or when not to create at all.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.