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

Annotations already cover read-only/non-destructive/openWorld, and the description adds real semantic context beyond them: 'state is attention or no_issues_observed, never healthy' and 'not_observed lists what this cannot see' frame the output honestly, and 'reported, not judged' clarifies that a missing change-stream event is not an error. 'Reads only; changes nothing' merely restates annotations, earning no credit here.

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 body is a single ~150-word run-on sentence that presents an entire output inventory as a comma-and-parenthetical dump rather than front-loading or structuring it. Because no output schema exists the return detail is justified, but the format is poorly scannable and the key purpose statement competes with the enumeration.

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 carries the return-value burden and does so thoroughly, covering state values, not_observed, budgets, review queue, and freshness vocabulary. What's missing is guidance on response size or when to fall back to granular sibling tools, but for an aggregate read tool this is largely complete.

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 100%, so both parameters (connection_id, days) are already documented, including the days default and range. The description alludes to 'runs that failed or stalled over the last days' but adds no format or semantic detail beyond the schema, so the baseline 3 applies.

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 opening clause ('Whether this connection's work is going well, in one read') is somewhat soft, but the enumeration of concrete signals (oauth_status, mirror freshness, run failures, review queue, budgets) makes the aggregate read purpose unmistakable. It also distinguishes itself from get_usage_summary by noting budgets are included 'as get_usage_summary reports it.'

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 ('in one read,' 'Reads only') but there is no explicit when-to-use-this-vs-a-granular-sibling guidance, e.g. whether to call get_operational_health instead of list_runs, get_readiness_scan, or get_usage_summary. The agent must infer that this is the aggregate overview tool.

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