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 discloses the subscription caps, the 60 s evaluation interval, edge-triggered re-arm behavior ('again only after it was false'), the 3-minute post-hoc delay for cluster_sweep, the one-time secret return, the exact HMAC verification formula and header name, and the auth tier requirement. These are precisely the behavioral traits an agent cannot infer from schema or annotations.

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 purpose, caps and the secret-return contract, then structured type-by-type, with zero filler sentences. It is a dense block of prose, but the density is justified by four distinct alert modes; light formatting or explicit per-type parameter listing would make it easier to scan.

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 mutation tool with no annotations and no output schema, the description covers what an agent needs to call it correctly: auth tier, payload contract, signature verification and type-specific parameters. Minor omissions remain, such as the response shape beyond the secret and whether/how the subscription is identified afterwards.

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 0% and there are 10 parameters, so the description must compensate, and it largely does: it defines the legal values for metric and op, gives a worked example for symbol/metric/op/threshold, and explains min_usd, x_normal and max_dist_pct in the context of their alert types. The gap is that it never states which parameters apply to which 'type' (e.g. that x_normal is only meaningful for cascade), leaving the agent to infer applicability.

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 a specific verb and resource: 'Subscribe an HTTPS URL to push alerts instead of polling', which immediately distinguishes it from the polling/query siblings in the list. It then enumerates the four alert types with concrete firing semantics, so an agent knows exactly what the tool produces.

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?

Gives clear per-type context for choosing among 'condition', 'cascade', 'hl_near_liq' and 'cluster_sweep', including defaults like 'no symbol = BTC, ETH, SOL' and the 60 s check cadence. It also flags the tier gate ('Needs a Pro or Bot key', 'Pro: 3, Bot: 20 subscriptions') and the 'instead of polling' rationale. It stops short of explicitly contrasting with siblings like get_liquidation_cascades, which would be the bar for a 5.

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