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

Annotations establish read-only/idempotent/non-destructive, and the description goes well beyond them: it discloses that call arguments are stored to improve recommendations, enumerates what is persisted (vendors, prices, seats, features, usage, billing) versus excluded (account keys hashed, billing/validation sources, IPs), notes results are summarized, and states exit tools log only counts/flags. This is unusually rich behavioral and data-handling disclosure.

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 the loop instructions are useful, but the description is a single dense block that mixes when-to-use, iteration mechanics, and a long retention/privacy disclosure. It also repeats the data-safety rule twice ('never raw exports, customer data, secrets or people's names' and again 'Don't send personal data or secrets'), which is redundant.

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 compensates by describing the return shape (missing facts ranked by stakes, each with where to find it, a status field, and next.tool/next.arguments) and the terminal condition ('enough'). For a multi-step tool with a nested manifest object, this is complete enough for an agent to drive the flow correctly.

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 already 100%, so the baseline is 3. The description adds real semantic value by explaining the relationship between parameters ('Send vendors and team_size (or a stack manifest)') and constraining values ('counts, prices, feature and job names only'), but it says nothing about decision or max_questions, so it does not fully rise to a 5.

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 a specific verb and resource (start a stack audit) and frames itself as the entry point: 'Start here when the user wants to cut spend across several tools, consolidate apps, or replace one tool safely.' An agent can tell it's the beginning of an iterative audit flow. It does not explicitly distinguish itself from the similarly named sibling audit_stack, so it falls short of a 5.

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 clear triggering conditions ('when the user wants to cut spend... and the bills alone won't decide it') and prescribes the full loop: send vendors/team_size, answer the returned questions, re-call until status is 'enough', then invoke next.tool with next.arguments. It stops short of naming when NOT to use it or pointing to audit_stack as the alternative, so it is not fully explicit about alternatives.

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