start_conversation

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

With no annotations, the description carries the full burden and does so richly: it discloses the conversational state machine, the auto-posted root task as turn 1, the turn accounting rule ('counts every automatic worker turn, result reviews included'), clamping ranges, the needs_human escalation path, and dedup idempotency returning duplicate=true. This is exactly the behavioral context an agent cannot infer from the schema.

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 purpose then behavior, and nearly every clause adds operational information. It is somewhat dense and the parenthetical '(goal, initiator = AGENT_ID, recipient, state active)' plus the mid-sentence line break cost a little readability, but there is little true waste.

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 six-parameter state-creating tool with an output schema and no annotations, the description covers lifecycle, limits, escalation, idempotency, and the sibling distinction. An agent has everything needed to invoke it correctly; only response_timeout_seconds is left to the schema.

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 0%, so the description must compensate, and it does for five of six parameters: max_turns (clamped 2..20, hard cap 20, counts all worker turns), expires_in_seconds (60..86400 s window), dedup_key (idempotent restart semantics), recipient (listener picks up), goal (posted as root task/turn 1). response_timeout_seconds is never mentioned, leaving one parameter's meaning entirely undocumented.

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?

The first sentence states a specific verb and resource — 'Start a bounded autonomous conversation with another agent' — and the closing sentence explicitly contrasts it with the sibling send_message ('An ordinary send_message never creates a conversation'). An agent can distinguish it from conversation, listener, and send_message without opening 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?

It gives clear triggering context ('so the recipient's listener picks it up without any manual phrase'), names the alternative that does NOT apply (send_message), and specifies when the conversation ends (completed, needs_human, turn limit, expiry, cancellation). No explicit 'prefer X when Y' for the other siblings like claim_message, 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.