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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuine output-shape context beyond annotations: findings are ranked by severity and player impact, each carries evidence plus a suggested fix, and a session property is named when most of a problem concentrates in one value. It says nothing about scan cost, coverage gaps, or how many findings are returned, which for a synthesis tool across 8 filters leaves real ambiguity.

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 first sentence runs well over 100 words, welding a dozen finding categories together with em-dashes and nested parentheticals; an agent must parse a wall of clauses to extract the core behaviour. The one genuinely differentiating line ('Start here when asked...') is buried at the end. It is information-dense but poorly front-loaded and unstructured.

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?

There is no output schema, so the description must carry return semantics, and it does: ranked, severity- and impact-ordered, evidence plus suggested fix, per-finding session-property attribution. It also enumerates the anomaly families sufficiently that an agent can predict whether its question is answerable. Minor gap: nothing on result volume or how to disambiguate when a finding points to another sibling.

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% (env, game, hours, props, service, version, session_type, steam_branch all documented inline), so the baseline is 3 and the description does not need to restate them. Beyond the schema, the description only reinforces the props behaviour ('naming a session property when most of a problem falls on one value'), without adding formats, defaults, or interactions between filters. Baseline 3 is correct.

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?

Opens with a specific verb+resource framing ('Ranked findings for a game') and enumerates the anomaly classes it detects, so an agent knows it is an aggregate diagnostic, not a single-metric reader. It does not cleanly separate itself from lookalike siblings (glitchless_performance, glitchless_network, glitchless_servers) whose domains it partly re-lists. The closing 'Start here when asked what's wrong with my game' sharpens the role, but only after a very long enumeration.

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?

Explicitly positions itself as the entry point ('Start here when asked...'), which is real when-to-use guidance and implies the siblings are drill-downs. It stops short of naming any sibling or stating when to skip triage in favor of glitchless_network or glitchless_get_error for a specific finding. No prerequisites or cost/latency guidance is given.

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