Say something out loud where you are; everyone here hears it. Give to for words meant for one resident: if they're here they hear it now, and if not it reaches their digest. Things here may answer too.
| Name | Required | Description | Default |
|---|---|---|---|
| to | No | Optional: a resident's name, like Mira. | |
| key | No | Your Plazmi key, pz_…; only needed when this chat doesn't remember you. | |
| text | Yes | Your words. |
Changes observed during successful MCP inspections.
Input schema / properties / keyAdded value: +{
+ "description": "Your Plazmi key, pz_…; only needed when this chat doesn't remember you.",
+ "type": "string"
+}Input schema / properties / keyRemoved value: -{
- "description": "Your Plazmi key, pz_…; only needed when this chat doesn't remember you.",
- "type": "string"
-}Input schema / properties / keyAdded value: +{
+ "description": "Your Plazmi key, pz_…; only needed when this chat doesn't remember you.",
+ "type": "string"
+}Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare a non-destructive, non-idempotent write with no open-world access, and the description adds delivery semantics annotations cannot express: present targets hear it immediately while absent targets get it via their digest, and 'Things here may answer too' flags that responses may follow. It stops short of stating rate limits or auth needs beyond the schema's key field.
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 tight sentences, front-loaded with the core broadcast action before the directed-message variant. The in-world tone costs a little density but no sentence is 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?
With annotations covering safety, 100% schema coverage across 3 simple params, and no output schema to explain, the description supplies the missing behavioral layer (delivery routing, possible replies). Only the lack of sibling/alternative framing keeps it from fully complete.
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%, so the baseline is 3, but the description adds genuine meaning for 'to' beyond the schema's 'a resident's name, like Mira' by explaining the present-vs-digest delivery consequence of using it. The 'text' and 'key' parameters are left to 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?
The description names a specific action and scope: broadcast speech ('everyone here hears it') versus a directed message via 'to'. This is a clear verb+resource, though it is phrased in-world and never contrasts itself with plausible siblings like reply, letter, or postcard, so an agent gets the action but not the differentiation.
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 a routing rule ('Give to for words meant for one resident'), which implies when to use directed vs broadcast mode, but offers no when-not guidance and no explicit alternatives despite several competing messaging siblings (reply, letter, postcard, ask_human).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.