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

Annotations indicate destructiveHint=true and idempotentHint=false, but the description explicitly says it is idempotent ('re-running is safe, never double-counts'), which contradicts the annotation. However, the description adds valuable behavior details: it reads paid Stripe customers (read-only), matches by email, records deals, and mentions the approval workflow. It also explains that it only works if Stripe is connected. The contradiction with idempotentHint is a minor issue, but it doesn't fundamentally mislead; the description actually clarifies a nuanced behavior. Since it adds significant context beyond annotations, I give a 4, but flag the contradiction.

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 description is quite long but information-dense. It front-loads the core purpose and key facts (idempotent, read-only, works only if connected). However, there is some redundancy with the title and routing, but every sentence adds value. The write-tier approval note is lengthy but necessary. It is structured in clear paragraphs, making it easy to scan. A score of 4 because it is longer than ideal, but each part has a purpose.

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?

Given the tool's complexity (mutation, approval workflow, idempotency, external dependency), the description covers all key aspects: what it does, when to use, prerequisites, approval flow, and idempotency. With no output schema, it mentions the effect (records closed_won deals) but not return format, but that is less critical. It is complete enough for an agent to decide and invoke 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?

The schema describes the single parameter (companyId) as 'FreedomOS company id to act within', which is standard. The description doesn't add much about the parameter itself, but it explains the broader context (needs company context). Since schema coverage is 100% and the parameter is self-explanatory, the description doesn't need to elaborate. It does add a note about membership requirement ('you must be a member'), which is helpful. Baseline is 3, and this adds a bit, so 4.

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 clearly states the specific verb ('Record won deals') and resource ('connected Stripe'), and it explains the purpose ('lead→paid conversion becomes measurable'). It distinguishes itself from siblings by focusing on Stripe conversions, whereas other tools like get_stripe_metrics only read metrics. It is a distinct action with a clear outcome.

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?

The description provides explicit guidance on when to use it: 'Use when conversion "isn't measured yet" or to refresh the conversion picture.' It also includes a routing path ('CRM/sales/revenue → measure conversion') which helps the agent navigate to the correct tool. However, it doesn't explicitly state alternatives, but the routing effectively implies the context.

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