Skip to main content
Glama

Dock

by ai.trydock

send_message

Send a direct message to another agent or human in the messaging substrate. Wires through cue.dock.svc, the same path the /live UI uses, so the recipient sees this message in their drawer (and, once they have a Dock-connected agent worker running, their agent harness's inbox). Address format is <agent_slug>@<user_slug>: flint@socrates targets the flint agent owned by user socrates; self@<user_slug> targets a human's synthetic self-agent (use this to message a human directly when you don't know which of their agents to ping). Use this to message a HUMAN (self@<user_slug>); to reach an agent teammate use message_teammate (send_message refuses agent recipients — see below). Don't use it as a chat-ops side-channel for things that belong in workspace events. Sender identity follows the caller: agent callers send AS themselves, user callers send AS their self-agent (self@<their_slug>). Body cap is 32,000 chars. Returns { messageId, threadId, to } on success. IMPORTANT: this tool QUEUES a message to the recipient's inbox — it does NOT wake or run a recipient agent. To actually reach an AGENT teammate (deliver AND wake it), use message_teammate. send_message to an agent teammate is REFUSED with an error (code -32602) — it would only file to their inbox without a wake, so the message would strand unacted-on. Use send_message ONLY for humans (self@<user_slug>) or when you deliberately want a human inbox drop without a wake. The recipient is resolved against the substrate's identity space, NOT against your accessible workspace set, this is messaging, not workspace write access. Pre-cue.dock.svc-deploy environments return cue_not_configured (caller treats as 'messaging not deployed yet').

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
toYesRecipient address in the form `<agent_slug>@<user_slug>`. Examples: `flint@socrates` (agent), `self@govind` (human's self-agent — use to DM a person directly). You may also pass an agent's stable id (`agt_...`) exactly as returned in `address_book`'s `agentId`. PREFER the id: an address is derived from the agent's slug, which does not follow a rename, so an address built from a teammate's current display name can fail to resolve. The id never changes.
bodyYesMessage text. Plain string, 1-32000 chars. `@<slug>` mentions inside the body CC the named agent on the message.
replyToNoOptional cue message id to thread under. When set, the recipient's drawer renders this as a reply with an inline parent-preview. Get the id from a prior `send_message` response or from the recipient's inbox listing.
send_atNoOptional ISO-8601 UTC timestamp to schedule the message for future delivery (e.g. `2026-06-04T15:00:00Z`). Omit to send now. A past timestamp is treated as send-now. Honored only where scheduled send is enabled; otherwise ignored. `scheduled_at` is accepted as an alias.

TDQS

A4.9/5.0
Behavior5/5

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

With no annotations provided, the description carries the full burden and succeeds: it discloses that the tool QUEUES without waking, refuses agent recipients with a specific error code, resolves against the substrate identity space, follows the caller's identity, enforces a 32k char cap, and returns a specific object. It also covers the pre-deploy 'cue_not_configured' error condition. No behavioral aspect is hidden.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long, but every sentence adds critical operational context (queuing semantics, refusal behavior, sender identity, error handling). It is front-loaded with the core purpose and flows logically from usage to constraints to return value. It could be slightly tightened, but no sentence is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a messaging tool with nuanced agent-vs-human behavior, the description covers every necessary aspect: API path, UI correspondence, address resolution, identity, message cap, return value, scheduling, error cases, and explicit alternatives. With no output schema, the description even provides the return shape. This is fully complete for an agent to select and invoke safely.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although schema coverage is 100%, the description adds substantial meaning beyond the schema: it explains the address format ('<agent_slug>@<user_slug>') with concrete examples, the 'self@<user_slug>' convention for humans, mentions body semantics ('@<slug>' mentions CC), replyTo rendering behavior, and the send_at alias. This far exceeds the baseline.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb ('Send'), the resource ('a direct message to another agent or human in the messaging substrate'), and the scope (address format, examples). It explicitly distinguishes itself from the sibling 'message_teammate' by noting it refuses agent recipients, making its purpose and boundaries unmistakable.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-to-use ('to message a HUMAN', 'when you deliberately want a human inbox drop without a wake'), when-not-to-use ('Don't use it as a chat-ops side-channel', 'send_message refuses agent recipients'), and names the alternative tool for agent teammates ('use message_teammate'). This is the gold standard for usage guidance.

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

All Dock toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.8/5.0
Disambiguation4/5

Most tools have clearly distinct purposes, with detailed descriptions that prevent confusion. The main ambiguity arises from send_message vs. the referenced but missing message_teammate tool, and add_column vs. update_surface for schema changes, but these are mostly clarified by the descriptions.

Naming Consistency4/5

The naming convention is predominantly verb_noun with underscores (e.g., create_workspace, list_rows, update_doc). Exceptions like 'search' and 'address_book' (no noun) and the two-word 'react_to_comment' are minor deviations in an otherwise consistent pattern.

Tool Count1/5

With 68 tools, the surface is far too large for an MCP server, exceeding the 50+ threshold for extreme mismatch. This volume creates excessive selection overhead for agents and suggests the tool set could be consolidated or split into focused servers.

Completeness3/5

The server covers broad functionality across workspaces, docs, tables, HTML, comments, files, webhooks, and billing. However, notable gaps exist: the explicitly referenced message_teammate tool is missing (preventing agent-to-agent waking), and there is no create/upload file tool or create API key tool, which creates dead ends in workflows.

Resources