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

With no annotations at all, the description carries the full burden and does add real behavioral facts: responses are capped near 12 KB and an unknown kind returns the list of kinds, plus default values (window=24 hours, limit=20) and detail=true semantics for peg_monitor. It omits the safety profile entirely -- nothing states whether the call is read-only, whether it requires authorization, or how truncation at 12 KB is signaled to the caller.

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?

Purpose and the 'pick a kind' instruction are front-loaded, and given zero schema documentation nearly every clause about entity formats and units earns its place. Structure is weaker than the content: the kind list is a comma-separated run-on with mid-entry line breaks (peg_monitor, record, shock), which slows scanning for a 17-item menu, though nothing is genuinely redundant.

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 dispatcher covering 17 heterogeneous kinds with 6 undocumented parameters and no output schema, the description supplies the entity contract, units, defaults, and volume cap that an agent needs to construct a valid call. What it lacks is any indication of return shape or truncation behavior per kind, which matters most for computational kinds like shock and record that produce derived verdicts rather than raw records.

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 0% and every property is a bare title, so the description must compensate -- and it largely does, defining entity type per kind (coin id, protocol slug, bridge id, exchange id, token contract address), entity syntax for record with concrete examples (sky_star:grove, morpho_vault:8453:0x...), units that change per kind (window in days, hours, or bps), the severity enum values SCO60|MICA67|GENIUS, and default limits. Gaps remain for kinds described only as 'entity=asset id' and for limit's meaning on kinds other than divergence/riskiest_wallets/scoring_backlog.

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 sentence states a specific resource (market, governance and risk data) and the description immediately enumerates 17 concrete 'kind' values, so an agent knows exactly what class of data it can pull. It stops short of a 5 because it never differentiates itself from siblings like entity_read, get_stablecoin_scores, or score_history, which plausibly overlap with kinds such as riskiest_wallets and scoring_backlog.

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 guidance is implied through the kind menu plus per-kind entity required/optional markers (e.g. 'yield_data (entity=protocol, optional)' vs 'liquidity_depth (entity=coin id)' with no optional flag), which helps pick a kind and know whether to supply entity. However there is no when-not guidance and no explicit routing against the six sibling tools, leaving the agent to guess whether a stablecoin question belongs here or in get_stablecoin_scores.

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