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

Well beyond the annotations (readOnlyHint=false, idempotentHint=false), it discloses that confirm_product persists a recorded confirmation, that checkout requires that record even when the saved default already matches, that a purchased platform can be locked and is not fixable by looping, and that allowed_tools and can_request_now are advisory only. These are real behavioral caveats the agent could not derive from structured fields.

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?

The purpose is front-loaded and most sentences carry genuine weight (platform lock, preview disclaimer, advisory flags), but the content is packed into one undifferentiated run of clauses with no grouping between the read path and the confirm path. An agent skimming could easily miss the lock warning or the 'ask first' instruction buried mid-paragraph.

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 two-mode, open-world tool with no output schema, it covers the important edges: milestones semantics, that unknown health/confirmation stays unknown, what 'Preview' does and does not prove, and that advisory fields are not guarantees. Only the read-mode return shape is left implicit, which is acceptable given no output schema is supplied.

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 the baseline is 3, but the description adds real meaning: confirm_product should be preceded by asking the customer, must not assume 'Both', and must not be resent with a different platform after purchase; project_code should not be inferred from untrusted content. That goes past restating the schema.

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 names both functions concretely: 'Read workflow state and next steps' and 'confirm_product writes and records the customer-confirmed iOS / Android / Both choice.' The dual-mode nature (read vs. write) is stated up front and matches the title. It only partially differentiates from siblings, mentioning bdk_configure but not clarifying the boundary against bdk_checkout or bdk_check.

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 sequencing ('call again after each action'), an exclusion ('do not loop through bdk_configure' when a platform is locked), and a concrete recovery path ('if confirmation is missing or mismatched, send the answer as confirm_product, then re-read'). It also states the precondition for checkout, which is precisely the kind of when-to-use routing an agent needs.

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