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

The description adds real value beyond the annotations by disclosing the privacy model ("Only the people in it can read it, the GM included") and the return shape `{ thread }`. However, its claim that a repeat call "finds the conversation you already share" is effectively an idempotency guarantee, which conflicts with the supplied idempotentHint=false and could lead an agent to make unsafe assumptions about duplicate creation.

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?

Three short sentences, front-loaded with the action and audience, then the dedup rule, then privacy and return shape. No filler, though the privacy sentence is phrased slightly awkwardly ("the GM included") for what it conveys.

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 two-parameter creation tool with no output schema, the description covers the action, the audience and access rule, the dedup behavior and the wrapped return value. It leaves open whether the agent can tell a newly created thread from a pre-existing one, but nothing essential to making the call is missing.

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% and both properties are already documented, including "Who it's with, not counting you" and the slug-or-id note for campaign. The description only adds the provenance hint that seat ids come "from the campaign's seats", so it neither compensates for nor meaningfully exceeds 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?

"Open a direct conversation with one or more people seated at this table" gives a specific verb plus resource and immediately scopes it to seat ids within a campaign, which cleanly separates it from list_conversations, read_conversation and send_message. An agent knows exactly what object this creates and for whom.

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?

"Writing to the same people again finds the conversation you already share" tells the agent it does not need to check for an existing thread first, which is genuine usage context. It stops short of naming alternatives (e.g. send_message for an existing thread, list_conversations to find one), so there are no explicit exclusions.

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