stats

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

With no annotations, the description carries the full burden and does so well: it declares read-only semantics, no model call, no caching, and per-view behavior nuances (provider view sees council members individually and circuit_open sheds). Minor omissions like empty-window behavior and bucket ordering prevent a 5, but the safety and side-effect profile is fully disclosed.

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?

Five dense sentences, front-loaded with the core purpose before any detail. Every clause earns its place: scope, bucketing/aggregates, filters, council-record extras, example queries, and the no-call/no-cache guarantee — no filler or repetition of schema content.

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 there is no output schema, the description compensates by naming the returned metrics (calls/allowed/refused/errors, avg and p95 latency, error_rate, totals, plus quorum/consensus/synth_fallback), which is the critical missing piece. It covers defaults, filters, and view semantics; the only gaps are the exact JSON response shape and bucket limits/ordering, which are minor for a read-only stats tool.

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 coverage is 100% with genuinely descriptive parameter docs, so the baseline is 3. The description adds interpretive value beyond the schema by explaining that model/session filters 'narrow to one backend or workflow' and by listing the metrics produced per bucket, which gives the agent a mental model of what bucketing by each key yields.

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 first clause states a specific verb and resource: "Read-only usage/health stats aggregated from the ask_fable audit log." It further distinguishes itself from the large ask_*/list_*/trace_* sibling family by framing itself as the performance-monitoring tool ('see how the tools are performing without spelunking JSONL'), which no sibling claims.

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?

Concrete example questions ('is GLM erroring a lot today?', 'how slow are councils this week?') tell an agent exactly what kinds of queries route here, and 'makes no model call and is never cached' clarifies when it's safe/appropriate to invoke. It does not explicitly name exclusions or alternatives (e.g., session_stats for per-session detail), so it stops short of full when-not-to-use guidance.

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