workflow_start

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

Annotations already declare readOnly=false, idempotent=false, destructive=false, openWorld=true, so the safety profile is partly carried. The description adds genuinely useful non-annotation context: async return with an operation ID, durability, target binding, middleware re-entry per step, and rejection of nested workflows/pipelines. It does not note idempotency/key semantics on retry, which is the notable remaining gap.

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?

Three labeled sections (Purpose/Safety/Returns) front-load the core action and keep the text short with no filler. The bolded section markers are slightly heavier than needed for a description this length, but nothing 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?

An output schema exists, so the brief returns note is sufficient and the description needn't detail the operation summary shape. For an async, multi-step tool it covers the essential behaviors, though it omits lifecycle specifics such as how long an operation remains queryable or retry behavior on the same target.

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 coverage is 100%, so both top-level parameters and their nested fields are already documented in the schema, making 3 the baseline. The description adds only conceptual framing ('target-bound', 'bounded steps') without new meaning about argument structure or constraints beyond what the schema states.

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?

States a specific verb+resource ('Start a durable, target-bound workflow') and the immediate outcome ('return immediately with an operation ID'), which is the key behavior distinguishing it from a blocking call. It also names the sibling tools (workflow_status, workflow_resume, workflow_cancel) that handle subsequent lifecycle stages, so an agent can place it precisely in the workflow family.

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?

Clear that this is the entry point and routes later phases to workflow_status/resume/cancel, plus a stated exclusion ('nested workflows/pipelines are rejected'). No explicit statement of preconditions (e.g. required profile/permissions) or when a plain tool call is preferable to starting a workflow, so it stops short of full when/when-not guidance.

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