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

Annotations already cover safety (readOnly, non-destructive, open-world, non-idempotent), yet the description adds substantial behavior the annotations cannot: oldest-first ordering, cursor semantics tied to the previous answer, freshness/access tiers, and the caveat that funds without an SEC CIK produce no filing events. It also defines the trigger condition for feed_stale/feed_paused (a Chainlink feed past its heartbeat while a session is open). Return shape is only partly sketched (id, at, discovery time), keeping it just short of 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 lead clause front-loads the resource and the event taxonomy well, but the rest is a dense chain of parentheticals and semicolon-joined clauses that is harder to parse than it needs to be. The 'Facts, not advice' closer and some tier detail could be trimmed without losing selection-relevant meaning.

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 three-parameter, no-output-schema feed tool, the description covers ordering, cursor pagination, filtering, freshness by tier, and a real coverage limitation (no CIK → no filings), plus a partial sketch of returned fields. An agent can call this correctly and interpret staleness without any other source.

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 100%, so the schema already documents kind, limit, and since, including the keyless-preview override — the baseline of 3 applies. The description's added value is modest: it explains what the kind values mean in the world (corporate_action = status changes, filing = 8-K/10-Q/10-K/Form 4/13D/G by acceptance time) and reiterates the 'since = last answer's next' cursor, but adds no syntax or edge cases beyond the schema.

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 description names a concrete resource and scope — 'What changed on Robinhood's Stock Tokens on Robinhood Chain' — and enumerates the exact event kinds (listed, delisted, multiplier_scheduled, feed_stale, filing, etc.), so the agent knows this is a change feed rather than a token lookup. It never names siblings like stock_token or stock_tokens, so differentiation from them is inferred rather than stated.

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 is implied rather than prescribed: polling for newer events via 'since' and narrowing via 'kind' are described, but there is no explicit 'use this instead of X when Y'. The tier rules (Team live, Pro 15 minutes late, keyless = last 10 events an hour old) do tell an agent when the answer will be stale, which is useful context but not a when-to-use rule.

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