check_health

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

The description extensively discloses behavioral traits: the tool can auto-repair (mutating, not read-only, consistent with readOnlyHint=false), some repairs are lossy and irreversible with caps, repairs are bounded per run and require re-running if `remaining` > 0, and unknown check names are rejected. It also explains the change in status reporting (healthy boolean removed), which adds context beyond the basic annotation.

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 information-dense and structurally sound, but it is quite long and covers many details that could be split into clearer sections. It front-loads the core purpose and health check list, but the later parts on fixes and status evolution are verbose. While every sentence adds value, the length might reduce readability for an agent scanning quickly.

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 (multiple checks, fix modes, cap limits, re-run conditions, status derivation), the description is thorough and covers edge cases like irreversible repairs and the pre-version status boolean. It does not have an output schema, so the description compensates by detailing return fields (`status`, `issues`, `severity_summary`, `remaining`, `checks_run`) and their meanings, making it complete for an agent to invoke and interpret results.

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?

While the schema describes each parameter briefly, the description adds significant meaning: `fix` is explained as auto-repair with lossy consequences, `checks` lists how to use it (subset, unknown rejection, `checks_run` echo), and `agent_id` is simply 'Agent ID' but the description implies agent-scoped behavior in the health checks. The description enriches understanding beyond the schema's basic descriptions.

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 specifies the verb ('check') and resource ('memory database health'), enumerates the specific kinds of issues detected, and distinguishes it from sibling tools like deep_check by detailing the exact health checks performed. It goes beyond a generic purpose by naming concrete checks and outputs.

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 the tool, including details on using the `checks` parameter to run a subset and the behavior of `fix=true` for auto-repair. It also clarifies the difference from the removed `healthy` boolean and how to interpret results, which helps the agent decide when to invoke this tool versus alternatives.

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