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

Annotations already declare readOnlyHint=false, destructiveHint=false and openWorldHint=false, and the description adds substantial context beyond them: three explicit refusal conditions (duplicate id, fromId==toId, unknown block) with 'nothing written', and the downstream effect of `shares` on compute_group_ownership with the ignored-value rules. It does not state what a successful response returns, but with no output schema that is a minor gap for a create tool.

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?

Purpose and the primary type distinction are front-loaded, and the later clauses on refusals and alternatives each carry decision-relevant information. The single dense paragraph duplicates the `shares` rules already present in the schema and would read better split into type/refusal/related-tool segments, but there is little true filler.

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 13-parameter mutation tool with no output schema in a domain-specific org-chart model, the description supplies the semantics an agent cannot infer: type meanings, automatic-creation paths, refusal behavior, and which sibling to use instead. The omission of `expectedProjectId`'s project-targeting behavior is the main remaining hole.

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 92%, so the schema already documents nearly every parameter, including the `shares` look-through rules that the description largely repeats. The genuine additions are conceptual: what 'business' vs 'structural' actually mean and how `style` interacts with the default dashed rendering. `expectedProjectId`, the only cross-project switching parameter, is not mentioned 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?

"Adds one dependency (arrow) between two existing org blocks" states a specific verb, resource, cardinality and scope. It immediately differentiates from the nearest lookalikes by naming add_relation (ERD field relations) and connect_kind_nodes (arch/process/journey/captable flows) as out of scope.

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?

Explicit guidance on which of the two types to use and when: 'business' for ownership/lease/service relations including parent-to-child stakes, 'structural' only where add_cell (parentId) and move_cell do not apply, e.g. hat-mode parents. It also names the alternatives (add_relation, connect_kind_nodes, update_dependency, delete_dependency) with the conditions that select them.

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