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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), and the description adds real operational context on top: filters apply before pagination, cursors must be cleared when filters change, epoch must be retained while paging observation, catalog metadata is unpaginated while other pages are bounded, and responses contain no source/credential dumps. It does not disclose auth/permission requirements or how conformance/run_services behave.

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?

Roughly 110 words with no filler, and the core capability statement is front-loaded ahead of the operational caveats. It is dense to the point of terseness, however — semicolon-chained fragments like 'Read build includeManifest or one operationId' are grammatical shorthand rather than clean prose.

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?

For a tool exposing ~19 distinct actions with no output schema and near-zero schema descriptions, the entry needs more disambiguation than it provides. Several actions (conformance, run_services, pending, observation, history, dataset) are only glancingly or not at all described, so an agent may not know what each returns, even though pagination and scope notes partially cover the shared mechanics.

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 effectively 0% across a large discriminated union, so the description must carry the load. It partially does: it explains the build action's includeManifest and operationId ('authoritative authored contracts') and touches filters, cursor, and epoch semantics. But the majority of variant-specific fields (rowQ, pendingQ, activityQ, namespace, activityKind, kind, order, source, bindingId, evidenceCursor) get no explanation, leaving clear gaps.

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 names specific verbs (discover/inspect/read) and enumerates the resource classes it covers: catalog metadata, acquisition receipts, builds/datasets, instances/history, and run-service evidence. That is enough to separate it from write-side siblings like firedrill_service_acquire or firedrill_service_publish, though it never names an alternative tool explicitly and leans on undefined jargon ('qualified', 'authored contracts').

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 routing advice among the tool's own actions: 'Use instance_history for newest instances with search/status/lifecycle filters; instances retains legacy ascending numeric pages.' That is genuine when-to-use guidance, but it offers no exclusions or comparisons against the other firedrill_service_* siblings, so the agent must infer read-vs-write boundaries.

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