Create a workflow. The workflow ID (UUID) is assigned automatically. For AI agent workflows, call get_agent_framework_docs first and use Mastra (bun) or LangGraph (python).
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | ||
| language | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral disclosure burden. It does reveal a useful hidden behavior—the workflow ID is auto-assigned as a UUID—but omits other key traits of a state-changing create operation: persistence, authorization needs, draft status, and behavior on naming conflicts. This is thin for a mutation tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences carry all the essential content: core purpose, auto-ID behavior, and framework-specific guidance. The main clause is front-loaded and no sentence is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a creation tool with no output schema, important context is missing: what the call returns, whether the created workflow is a draft or executable, and how it relates to update_draft and publish_workflow. The framework prerequisite is helpful, but an agent still has to guess at the workflow lifecycle and return value.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It adds real semantic value by mapping the language enum values to frameworks: bun means Mastra and python means LangGraph. It provides no guidance for the required 'name' parameter, making the compensation partial.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description opens with 'Create a workflow', naming a specific verb and resource, and is distinguishable from siblings like get_workflow and publish_workflow by its creation intent. It does not, however, explain whether the created workflow is a draft or immediately usable, which would further sharpen the purpose.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It provides explicit usage guidance: for AI agent workflows, call get_agent_framework_docs first, then choose Mastra (bun) or LangGraph (python). This tells the agent when to consult a sibling and how to select the language parameter. It does not state when create_workflow should not be used, such as when updating or publishing an existing workflow, but the primary routing advice is strong.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Every tool has a clearly distinct purpose: workflow lifecycle, invocation, logging, metrics, KV store, secrets, connections, and documentation. Even similar tools like get_logs vs get_invocation are well-differentiated by descriptions.
Most tools follow a consistent verb_noun pattern (create_workflow, get_workflow, list_workflows). The KV tools (kv_get, kv_set, kv_list, kv_delete) are internally consistent but deviate from the dominant verb_noun style by using a noun_verb prefix.
With 18 tools, the count is slightly above the typical 3-15 range, but the broad platform scope (workflow management, invocation, logging, metrics, KV, secrets, connections, docs) justifies each tool's existence. No tool feels redundant.
Workflows have create, read, update (via update_draft), list, and publish, but lack a delete operation. Secrets support create and list but no delete/update, and connections only have list. These lifecycle gaps create potential dead ends for agents.
