yieldsentinel

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

Annotations already establish readOnly/idempotent/non-destructive, and the description adds genuine context beyond them: it fetches live yieldhunter research, returns a PolicyVerdict plus A2H (summary, suggestedNextSteps, legend), degrades to review/unknown on missing data rather than a silent ok, and explicitly states it never holds keys and is not an exit/execution signal.

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 purpose and the critical 'no policy = no verdict' warning, and the example is useful. However the preset list is stated twice and 'Scoring ≠ policy approval' appears in both the description body and the requireCurated text, which is avoidable duplication in an already dense block.

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 an 18-parameter tool with no output schema, the description covers the input contract (required policy inputs, id alias resolution), the return payload (PolicyVerdict + A2H fields), and the failure semantics (review/unknown). Nothing an agent needs to call this safely and interpret the result appears 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?

Schema coverage is 100% so the baseline is 3, but the description goes further by explaining that poolId/opportunityId/id are aliases for one market and by naming the policy inputs (policyPreset, minScore, maxRewardShare, excludeCategories, maxExitRisk, requireCurated) that are needed for a real check despite schema required=0. It adds less on the scoring/notional params, which the schema already covers.

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 sentence gives a specific verb and resource ('Evaluate a described stablecoin yield position against a research policy') and explicitly frames it as a pull. It's instantly distinguishable from the only sibling, 'health'.

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 states a hard prerequisite ('Always pass your policy rules... Without a policy... the verdict will be review/unknown, never ok') and enumerates the valid policy inputs. It also routes the agent on which identifier to use ('Pass poolId from yieldhunter list_yields opportunities[].id... Prefer poolId... when the protocol has more than one market').

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

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

Annotations already cover read-only/idempotent/closed-world, yet the description adds meaningful behavior: the server is stateless, the X-YHYS-Test header must be sent on every request, and the ping may refresh research vintage (a side effect for a read tool). The toolStats/toolStatsExternal/toolStatsInternal and clientInfo.name prefix sentences are opaque and add bulk without clarifying invocation behavior, keeping this below a 5.

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 purpose and the call-to-check_position guidance are front-loaded, which is good. But the middle block about toolStats buckets and yhys-qa/yhys-feel clientInfo prefixes is dense internal jargon that consumes attention without helping an agent decide to call or how.

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?

With no output schema, the description carries the return-value burden and does list the main response fields (status, version, schemaVersion, policyVersion, feature flags, tool list) plus the stats buckets. It stops short of describing field shapes, but covers enough for a zero-param health probe.

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?

Zero parameters, so the baseline of 4 applies. The only invocation-relevant detail is the required X-YHYS-Test header, which is disclosed in prose even though it isn't in the schema (additionalProperties: true).

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+resource ("Liveness and capability check for yieldsentinel") and enumerates what it returns: status, package version, schemaVersion, policyVersion, feature flags, and available tools. It explicitly separates itself from the sibling by noting "a verdict still requires check_position."

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?

"Use this first to verify connectivity, then call check_position with policyPreset treasury-conservative | balanced-research | rwa-selective" gives an explicit ordering, the alternative, and even the downstream parameter values. The caveat "YH healthy ≠ YS ok" tells the agent this is a precondition check, not a verdict.

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