quality_report

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

With no annotations, the description does real work: it discloses that results are "Computed entirely from live DB queries — no LLM involvement," which tells the agent the call is deterministic, costs no model spend, and is safe to repeat. It also names the underlying signals (recorded llm_spend, curation precision with its denominator) so the agent knows facts are derived, not estimated. It stops short of stating read-only guarantees, freshness/latency, or whether the snapshot is expensive to compute — the main remaining gaps.

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?

Front-loaded with the one-sentence purpose, then a structured enumeration of contents with backticked field names, ending with the cost/disclosure caveat. It is dense but readable and every clause identifies a distinct metric family. Slight redundancy: since an output schema exists, spelling out the returned fields in prose partly duplicates structured data.

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 zero-parameter, read-only reporting tool whose return shape is already captured by an output schema, this description covers what the agent still needs: what dimensions are reported, that the numbers come from live DB queries with no LLM cost, and what "precision" is measured against. Nothing required to decide to call it or to interpret the headline metrics is missing.

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?

The tool takes zero parameters, so there is nothing for the description to disambiguate; baseline 4 applies. One minor wrinkle: it says precision is measured "over the default window" even though no window argument exists, which could momentarily suggest a parameter that isn't there.

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?

States a specific verb ("Return") and a specific resource ("extraction-quality dashboard snapshot"), then enumerates the exact metric families contained (particle counts by status/calibration_source/schema_version, structural-mix percentages, llm_spend, curation_precision). It is unmistakably distinct from data-access siblings like particles_list, query, or lint — an agent can tell this is the read-only quality aggregation endpoint without opening any schema.

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 only implied: the enumerated contents make it obvious this is the tool for quality/health monitoring rather than data retrieval, but there is no explicit "use this when you need X instead of Y" statement or exclusion against siblings such as query or particles_list. Adequate for routing a competent agent, but the selection rule is left to inference.

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