Fetch Hive

Add a step to a workflow

add_step

Idempotent

Add one step: a step_type (get_step_type for its settings) or an app action (app_action from search_actions, with settings.integration_arguments from get_action_schema). Without placement it goes just before End; in a workflow with conditions or loops pass placement. A prompt step needs messages. The change is saved as a version: the answer's undo.version_id restores the steps from before it. Check problems in the answer before running.

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": {
      "workflow_id": "<workflow_id>"
    },
    "name": "add_step"
  }
}

Input Schema

Name	Required	Description
`name`	No
`messages`	No	Prompt steps only: system message, user message, then a user message of labelled placeholders.
`settings`	No	Field names and values from get_step_type (for an app action: integration_arguments, and integration_authorization_id when the app has more than one connected account).
`placement`	No	Where it goes, relative to step_id: after_step, condition_true or condition_false (first in that branch), after_condition, iteration_body (inside the loop), after_iteration.
`step_type`	No	From list_step_types. Leave out for an app action.
`app_action`	No	An action slug from search_actions, e.g. "SLACK_SEND_MESSAGE".
`workflow_id`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 10, 2026

TDQS

A4/5.0

Behavior4/5

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

Annotations already declare the safety profile (non-readonly, non-destructive, idempotent), so the description correctly spends its words on non-obvious behavior instead: the default insertion point before End, that the change is saved as a version, and that undo.version_id in the response restores the prior steps. It also warns to check reported problems before running. Permissions or authorization requirements beyond integration_authorization_id are not covered.

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?

Dense but front-loaded: the core action and its two modes come first, then placement, then messages, then versioning. The heavy parenthetical quoting of sibling tool names makes it slightly hard to scan, but no sentence is filler.

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 7-parameter mutation tool with nested objects and no output schema, the description covers the essential gaps: insertion defaults, mode selection, required prompt payload, and the fact that the response carries undo.version_id and problems. It stops short of describing the shape of the returned problems or authorization preconditions for app actions.

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

Parameters4/5

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

Schema coverage is 71%, and the description adds selection logic on top of it: step_type versus app_action is mutually exclusive by mode, settings come from get_step_type while integration_arguments come from get_action_schema, and messages are prompt-step-only. The detailed placement and messages ordering semantics largely restate what the schema already says, so the value added is real but modest.

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

Purpose4/5

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

States a specific verb and resource ('Add one step') and immediately splits it into the two supported modes, a step_type or an app action, so an agent knows what gets created. It does not explicitly distinguish itself from edit_step, move_step, or remove_step, so the sibling-differentiation bonus is only partial.

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 conditional guidance: omit placement for the default (just before End), pass placement when the workflow has conditions or loops, and supply messages for prompt steps. It also routes the agent to get_step_type, search_actions, and get_action_schema for the values to use. Missing only explicit when-not-to-use-this-tool guidance versus edit_step/move_step.

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

All Fetch Hive tools Inspect

Try in Browser

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?