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

Annotations already declare destructiveHint=true, openWorldHint=true and idempotentHint=false, so the safety bar is partly covered. The description still adds genuine substance beyond them: execution requires live authority plus separate approvals, discover and planOnly do not execute agents, retries should reuse a stable clientMessageId/assistId, and it flags a feature-gated local preview. It does not disclose what stop/pause/release_budget actually do to in-flight work, which is the main residual gap.

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?

It is a single dense, front-loaded paragraph with no filler sentences; the primary use case and routing rule come first, then constraints. The semicolon-chained style packs several distinct concerns (modes, idempotency, execution authority, preview gating) into a run-on that is slightly harder to scan than it needs to be.

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?

Output schema exists and the field-level schema descriptions are complete, so return values and parameter details need not be repeated. But for a dispatcher covering fifteen command variants and budget/approval flows, the description only explains the `turn` path; an agent must reverse-engineer quote, authorize_budget, execute, accept_step, stop and pause from the schema alone. Adequate for the common path, incomplete for the tool as a whole.

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 baseline is 3. The description still adds discriminating semantics the schema does not spell out as clearly: mode should be omitted for automatic routing, planOnly yields a proposal without execution, assistId is required for follow-ups, and clientMessageId must be stable across retries (that field has no schema description at all).

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 scopes the tool as "continue a task with a turn in discover, coordinate, or chat," which is a concrete verb+resource for the `turn` variant. But the schema exposes fifteen command variants (create, revise, quote, authorize_budget, execute, stop, pause, message, etc.), none of which are mentioned, so the stated purpose covers only a fraction of what the tool does. It also never distinguishes itself from the sibling `assist` or `act` tools.

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 gives real selection guidance for the mode axis — omit mode for auto-routing, discover/planOnly do not execute, chat does not start discovery — which is useful when-to-use information. However there is no guidance at all on when to pick one of the other command variants (e.g. quote vs execute vs authorize_budget), and no guidance on when to use this tool versus the sibling `act`/`assist`.

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