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

The annotations already flag a destructive, open-world write, but the description adds governance semantics the annotations cannot convey: agent-created workflows get safe automatic trials by default, a person can disable that, and updates containing new writes still require human approval. It also discloses the hard cap that automatic share is at most 10%, which is a real behavioral constraint on the call.

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?

One dense but front-loaded paragraph: purpose first, then eligibility gates, then exclusions, then the follow-up tool, then limits and a concrete example. Every clause carries information the agent needs (the tie-fails rule and the save_version overlap are both non-obvious), and nothing is redundant with the schema or annotations.

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?

With no output schema, the description correctly tells the agent that this call does not return the decision and to follow up with experiment_status, and it covers the approval model and share ceiling. The remaining gap is the unexplained experimentId and the fact that the description never states what, if anything, comes back immediately from this call.

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 carry parameter meaning, and it only partially does. The example {"workflowId":"invoice-flow","share":5} plus "automatic share is at most 10%" give useful semantics for workflowId and share (percentage of traffic, ceiling of 10), but experimentId is never mentioned and no parameter is marked required, leaving the agent to guess when to supply an experimentId versus a workflowId.

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?

"Ask the server to start a limited trial" names a specific verb (start) and resource (a limited trial for a saved workflow version), and the description immediately scopes it against workflow_save_version, which auto-starts trials for in-chat updates. An agent can tell this apart from siblings like experiment_status or workflow_dry_run without opening any schema.

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 gives explicit preconditions (saved version not already in a trial, paired dry run must score higher than the active version, a tie fails, offline and shadow gates passed) and explicit exclusions (do not use to bypass review, do not use for an update in this chat because workflow_save_version already starts that trial). It also routes forward to experiment_status for the decision and metrics.

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