Read Engine Keyspace

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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=true, so the safety profile is covered structurally; the description still adds real value by bounding cost (2,000 keys, two seconds), disclosing that LRU/LFU clocks are not touched where the server allows it, and warning that key names may contain sensitive data and require the developer role. That auth/permission disclosure is exactly the kind of context annotations cannot carry. It stops short of full transparency because it says nothing about failure modes, cluster/version limitations, or what 'where the server allows it' implies for the caller.

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?

A single front-loaded sentence opens with the sampling action and its cost bound before covering grouping, output fields and the permission caveat. Every clause carries information, though the sentence is long and slightly run-on; splitting the permission note would improve readability without adding length.

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 spends words enumerating what comes back (memory share, key counts, TTLs, largest keys), and it discloses the sampling ceiling and the developer-role requirement. For a read-only inspection tool that is close to enough. It is not fully complete because the five context parameters remain unexplained and there is no note on behavior when the service lacks a database 0 or restricts clock-preserving scans.

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% across five parameters, so the description must carry the burden and largely does not: workspaceSlug, appSlug, serviceName, env and region get no explanation of expected format, defaults, or effect on behavior. The only parameter-adjacent fact is that database 0 is fixed, implying there is no database-index argument, which is a small but genuine clarification. This is insufficient for a tool with a completely undocumented 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 gives a specific verb and resource with hard scope: sample up to 2,000 keys of database 0 and group them by the prefix before the first colon, reporting memory share, counts, TTLs and largest keys. An agent immediately knows this is a read-only keyspace survey rather than a key fetch. It does not name or contrast with the closest sibling (getVolumeKeys) or the engine-inspection tools, so it falls short of the top mark.

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?

Usage is implied by the sampling/grouping behavior — an agent can infer this is for aggregate key-distribution inspection rather than retrieving a single key. However, there is no explicit when-to-use statement, no exclusion of alternatives such as getVolumeKeys or readEngineStorage, and no guidance on running it against production versus preview environments despite an env parameter.

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