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

Annotations only cover the safety profile (readOnlyHint=false, destructiveHint=false, openWorldHint=false); the description supplies the substantive behavior an agent cannot infer: writes require targeting the new project (expectedProjectId or set_active_project), nothing is written during the turn in the in-app chat, the sub-view inherits the host's rights and shares, and the call counts against the project cap. This is unusually rich disclosure for a mutation 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?

The core action is front-loaded in the first sentence, and every sentence carries distinct information (targeting, refusal, rights inheritance, alternative tool). It is long and heavily emphasized with capitals and parentheticals, which costs some readability, but little is genuinely redundant.

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 complex, side-effecting creation tool with no output schema and a 100%-covered input schema, the description covers the remaining risk surface: refusal semantics, rights/share propagation, where the write actually lands, and the chat-vs-MCP targeting difference. An agent has everything needed to call it correctly or route elsewhere.

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, but the description adds genuine meaning: primaryKind is framed as the place where cross-kind inception happens ('a BPMN process under an org chart block, an ERD under an architecture component'), and it clarifies that cellId is the host block that stays in this project while only a pointer is set. These reinforce rather than merely repeat 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?

States a precise verb+resource+effect: 'creates an INDEPENDENT (empty) project and links it immediately to the host block, in a single action.' It explicitly contrasts itself with the sibling that handles the already-existing case ('A project that ALREADY EXISTS is linked with link_project_to_cell') and implicitly with create_project, so an agent can separate it from neighbors without opening any 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?

Gives an explicit when-to-use rationale (the 'inception' pattern: keep the summary on the main view, push detail one level down) and an explicit when-not/alternative (link_project_to_cell for pre-existing projects). It also states the refusal condition — 'Refuses if the block already carries a link (no project is created then)' — which is exactly the routing information an agent needs.

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