send_message

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

With no annotations, the description carries the full behavioral burden and does so well: it discloses that turns on one session never overlap, that a message to a busy session blocks until the turn ends, that background mode changes the return shape and requires wait_thronglet to collect, that steer interrupts the running turn and jumps the queue, and that non-resumable sessions reject all messages including steer. These are exactly the traits 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?

Dense but front-loaded: the core purpose and the run_thronglet return-equivalence come first, followed by the blocking rule, then background, then steer, then the terminal limitation. Every clause carries information; there is no filler or restatement of the name.

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 6-parameter mutating tool with nested schema input, no annotations, and no output schema, the description covers the hard parts: concurrency, interrupt semantics, async collection, and the dead-session case. Minor gaps remain — nothing on error/timeout behavior and only a pointer ('same JSON as run_thronglet') for the default return payload — but nothing critical to correct invocation 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 description coverage is 100%, so the baseline is 3, but the description adds real semantics beyond the schema for two parameters: it explains what background:true actually returns and how to retrieve the result, and clarifies steer's queue-ordering effect relative to pending messages. session_id's provenance is also stated. timeout_s and schema remain schema-only, which keeps this from a 5.

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 specific verb+resource ('Sends the next message into an earlier session') and immediately anchors it to the sibling it depends on ('session_id from run_thronglet'), which cleanly separates it from run_thronglet (start a session) and wait_thronglet (collect a result). An agent can pick this over its siblings without opening a 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 explicit conditional guidance: use background:true when you want an early {session_id, state, queued} return and then wait_thronglet, use steer:true when you must interrupt the in-flight turn, and note that sessions showing accepts_messages: false in list_thronglets cannot receive messages at all. It does not spell out the inverse case (when to prefer run_thronglet over this), so it falls just short of fully explicit alternatives.

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