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

Annotations already declare the mutation/idempotency profile, and the description adds genuinely non-redundant behavior: the proposal is *pending* and human acceptance is a separate step, retries are disambiguated by exact identity plus unchanged body, and the response is a revision-bound command result with no read attestation or human acceptance. These are exactly the operational caveats an agent needs for a destructive, idempotent write. It stops short of describing failure modes (revision conflict, snapshot mismatch).

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?

Four compact sentences, front-loaded with the create action and its binding requirement before the identity/retry mechanics. Dense domain jargon ('revision-bound command result', 'read attestation') but no filler. Slightly overloaded in the final sentence, which packs return shape and two negative guarantees together.

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 destructive, idempotent mutation with a deep nested schema and no output schema, the description covers the essentials (pending state, human acceptance separation, retry identity, return shape). It omits what a caller most needs at the edges: what preconditions cause rejection (stale baseRevision, snapshot mismatch) and what the nested basis/recommendation fields are for. Adequate but not complete for this complexity.

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?

With 50% schema description coverage and a heavily nested command object, the description does useful work by naming the required identity fields (clientId, baseRevision, purposeRevision, snapshotId) and by framing supportsActionId/replacesDecisionId as 'supporting or replacing a pending choice'. It says nothing about basis, recommendation, briefExcerpts, or decisionText length constraints, leaving the largest part of the payload undocumented outside the schema. Baseline 3 for moderate coverage with partial compensation.

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 and resource: creates a *pending* decisionText proposal bound to a participant's earlier retained messageNumber, with optional support/replace behavior. That is materially more precise than the title alone and separates it from read-side siblings. It does not, however, explicitly contrast itself with close siblings such as propose_decision or purpose_support_decision, so an agent must infer the boundary.

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 stated: the proposal must be bound to an earlier retained messageNumber, and it can optionally support or replace a pending choice. There is no explicit 'use this instead of X when Y' guidance despite several plausible alternatives (propose_decision, purpose_support_decision, purpose_amend_decision) sitting in the sibling list.

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