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

Annotations already declare destructiveHint=true, idempotentHint=true and readOnlyHint=false, so the safety profile is covered structurally. The description adds real substance on top: the subset-versus-complete-set rule per kind implicitly warns that unlisted records can be displaced, and it states hard preconditions (parentId 'must not be ready' for work; instructions accepted only in the self-improvement stage). It does not describe failure modes or whether reordering is reversible.

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 a single-sentence statement of the operation, then a bulleted list where each bullet carries distinct kind-specific semantics rather than filler. It is long, but the length is dictated by nine enum variants that each behave differently. Minor repetition of the 'parentId is the project' pattern across bullets is the only slack.

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 nine-variant mutation tool with no output schema, the description covers what an agent needs to construct a valid call: the orderedIds contract, per-kind parentId, completeness expectations, and the workflow-stage restriction. Return values are not needed since the agent only needs to know the resulting order, which is implied.

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, but the description genuinely adds meaning beyond it. The schema for parentId explicitly defers to the description ('the tool description says what each kind takes'), and the description discharges that with a kind-by-kind mapping; it also clarifies orderedIds as a first-shown-first sequence with subset semantics.

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 plus the ordering semantics: 'Set the manual order of one kind of record, first shown first.' No sibling tool does ordering, so the agent can immediately distinguish it from the update_*/delete_* family. The 'one kind of record' framing scopes it precisely to the enum in the schema.

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?

The per-kind bullets give strong conditional guidance: which kinds need a parentId, which must be an exhaustive set ('every active Idea', 'every planned Batch') versus which tolerate a subset ('a subset keeps every other project in place'), and the stage constraint on 'instruction' during a workflow run. It never names an alternative tool or states when not to reorder, so it falls short of a 5.

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