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

Annotations only declare readOnlyHint and openWorldHint; the description supplies the material behavioral facts: figures are quantities not charges, network traffic is measured at container interfaces and overstates egress, memory peak is a 30-second sample not a true max, and null vs zero semantics. It also states the tool cannot mutate class and warns that no other tool or dashboard screen can — highly relevant for avoiding a false remediation path.

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 core purpose is front-loaded and every clause carries real information (sampling caveats, null semantics, parameter defaults). It is on the long side for a three-parameter read tool and the all-caps class-immutability warning is repeated for emphasis, but little of the text is filler.

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 and annotations that say nothing about returned data, the description carries the full burden and does so by explaining what each notable field (`notes`, `coverage`, null figures) means and how to interpret the numbers. An agent could call and correctly report this tool without further documentation.

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 0%, so the description must compensate and largely does: it enumerates the allowed `window` values (1h, 6h, 24h, 7d, 30d) and its 24h default, plus the `per_service` default of false. `project_id` is only implied by 'this project' rather than explained, which keeps it from 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?

First sentence states precisely what is returned — CPU, memory, disk, files and network consumption for a project over a window — with a specific scope qualifier ('what this project actually consumed'). This clearly separates it from siblings like get_project_detail, get_logs or get_events, which serve different purposes.

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 concrete operating context: read `notes` before restating figures, `coverage` tells you how complete the window is, a null is not a zero, and it prescribes what to tell a user when a project has outgrown its class. It lacks an explicit 'use this instead of X' comparison to sibling tools, so it stops short of a 5.

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