wait_for_notificationLong-poll for the next ping. Blocks until one arrives or the timeout elapses — the real-time inbound channel for an agent. The agent's own sends are excluded.
| Name | Required | Description | Default |
|---|---|---|---|
| after | No | Cursor from the previous call; omit to get the current head. | |
| timeout | No | Seconds to hold the request open (server-capped). |
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Output schema / properties / result / properties / notifications / items / properties / origin_agentAdded value: +{
+ "description": "Present when a connected agent originated the ping: its handle and display name.",
+ "type": "object"
+}Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish readOnly, idempotent, and non-destructive traits. The description adds genuinely useful behavioral context beyond those: it explicitly says the call blocks until a ping arrives or the timeout elapses, that it is the real-time inbound channel, and critically that the agent's own sends are excluded. This helps the agent predict runtime behavior without contradicting the annotations.
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 description is a single, tight sentence that front-loads the core action ('Long-poll for the next ping') and packs essential behavioral details into a short em-dash clause. There is no fluff, no repetition of schema field names, and every phrase earns its place.
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?
Given the output schema exists, the description does not need to explain return values. The key operational facts—blocking behavior, timeout elapse, inbound direction, exclusion of own sends, and real-time nature—are all present. Combined with the schema and annotations, an agent has everything needed to invoke this tool correctly.
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%, meaning both 'after' and 'timeout' are fully documented in the schema itself. The description provides no additional parameter-level guidance beyond what the schema states, so the baseline 3 is appropriate. It does not hurt, but it also does not compensate or enrich beyond the structured 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 uses a specific verb phrase 'Long-poll for the next ping' and immediately defines the resource being waited on. It also adds scope ('the real-time inbound channel for an agent' and 'own sends are excluded'), making it easy to distinguish from siblings like get_notification or list_notifications.
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?
The phrase 'real-time inbound channel for an agent' clearly communicates the intended use case: blocking to receive incoming pings. It does not explicitly name alternatives or conditions for when not to use it, but the blocking semantics versus the polling nature of get_notification/list_notifications is strongly implied. A small explicit comparison would have made this a 5.
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.
Several tools overlap in purpose: broadcast, trigger_quick_action, create_handoff, ask_question, request_approval, and live_status can all deliver a ping or prompt to a human, and the wait_for_* family has five similar long-polling variants. The detailed descriptions help, but the update_quick_action/update_quick_actions pair and the activate_agent_inbox/handoff flow still create real selection ambiguity.
The set overwhelmingly follows verb_noun conventions like create_, get_, list_, update_, delete_, and wait_for_, making the naming predictable and searchable. A few outliers — broadcast, connection_info, and live_status — break the pattern, and the live_status/get_live_status pair is less obvious than it could be.
40 tools is a heavy surface for one MCP server, even with a broad feature set spanning rooms, webhooks, quick actions, questions, approvals, handoffs, attachments, and live status. The count feels inflated by many narrow wait/get variants that could be consolidated or exposed through fewer, more composable operations.
Core workflows around questions, approvals, handoffs, webhooks, quick actions, and attachments are mostly covered with create/read/list/wait operations. However, the room lifecycle is missing obvious delete or leave operations, and approvals lack list/cancel capabilities, leaving notable gaps for an agent managing those resources.