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

Annotation Contradiction: the description documents clear mutations — register creates a persistent intent, delete removes one — yet annotations declare readOnlyHint=true and destructiveHint=false. That directly contradicts the register/delete semantics. Per the rubric, score is 1 and the flag is set. Notably, in isolation the description itself is exemplary (HMAC signing, SAVE-the-secret warning, silent watermark init, 5-straight-failure auto-disable, ~2h evaluation, HTTPS-only host) — the failure is the conflicting annotation, not the text.

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 description is long but earns its length for a complex 7-parameter tool — every clause carries operational info. The core push-vs-poll concept is front-loaded before the parameter rundown, and the register response/secret warning is placed before the failure semantics. The density is high and would benefit from light formatting, but there is no waste given the tool's complexity.

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 complex tool with 7 interdependent parameters, an output schema, and critical security requirements, the description is essentially complete: it covers the action/kind matrix, webhook URL constraints, HMAC signature format, secret lifecycle, watermark initialization, failure/disabling behavior, evaluation cadence, and correct sibling routing. Return values are covered by the existing output schema. Nothing an agent needs to call it correctly is missing — the sole blemish remains the annotation contradiction, handled above.

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. The description adds real value beyond the schema: it maps each kind to the relevant watch parameter ('new_deal_in_market' watches deals in params market, 'news_keyword' watches news matching q, 'permitting_change' optionally per state), clarifies webhook_url is public-HTTPS-only with private/internal hosts rejected, and explains intent_id's role for delete. This cross-parameter relationship matrix is exactly what the schema alone does not convey.

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?

Opens with 'STANDING QUERIES with webhook push — register an intent once and DC Hub POSTs an HMAC-signed webhook...' giving a specific verb, resource, and mechanism. The push-vs-poll contrast and natural-language intent examples ('notify my orchestrator on any new deal in Columbus') make the purpose unmistakable, and the final sentence names the sibling alternatives (get_news, list_transactions, set_market_alert) it is not meant to replace.

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?

Provides explicit when-to-use framing ('push, not poll', 'Answers "notify my system..."') and an explicit when-not-to with named alternatives: 'Do NOT use for one-shot reads (use get_news / list_transactions) or email alerts (use set_market_alert); this is machine-to-machine push.' Also includes a runnable try example selecting kind, q, and webhook_url. Nothing 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.