Taskriz

Get Replies

get_replies

Read-only

The Agent’s unacknowledged events, oldest first: replies (the answer as JSON in reply, the request in in_reply_to), requests people started from its catalog (initiate), and what became of email invitations (invitation: accepted — with the delivered request_id — declined or expired), and people who paired with its code (paired: their email and connection_id). They stay until you acknowledge them — call acknowledge once handled. Read receipts (receipt: delivered — it reached their device — then read, per person) are left out unless include_receipts is true. Optionally wait up to 20 seconds for something to arrive; call again to keep waiting. Give ONE of: thread (from send_request) — only that request's events; use it whenever you're following up a request — or all: true — every event, only when the person asks for all of the Agent's replies, or to see who paired or started a request. Acknowledge each reply once you've told the person about it.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "agent_id": "<agent_id>"
    },
    "name": "get_replies"
  }
}

Input Schema

Name	Required	Description
`all`	No	Every event, not just one request's — only when the person asks for all replies, or to see pairings and started requests.
`thread`	No	Only this request's events (its thread, from send_request).
`agent_id`	Yes	The Agent (from list_agents or create_agent).
`wait_seconds`	No	Wait up to this long (0–20) if there is nothing yet.
`include_receipts`	No	Also list read receipts (delivered / read, per person) — to see whether someone has had or opened a request.

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 6, 2026

TDQS

A4.3/5.0

Behavior4/5

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

Beyond readOnlyHint=true, the description discloses persistence semantics ('they stay until you acknowledge them'), ordering (oldest first), default exclusion of receipts, and the long-poll behavior (wait up to 20 seconds, call again to keep waiting). The one gap is what happens if neither `thread` nor `all` is supplied, even though it says to give ONE of them.

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?

Content is front-loaded with what is returned before usage rules, and nearly every clause carries information. The opening sentence is a dense run-on with nested parentheticals that is harder to parse than necessary.

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

Completeness4/5

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

With no output schema and readOnlyHint as the only annotation, the description compensates well by describing the returned event shapes and the acknowledge lifecycle. It stops short of covering the neither-parameter case and any pagination/volume behavior, which matter for an event-listing tool.

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

Parameters3/5

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

Schema coverage is 100%, so the schema already defines all five parameters; the description mostly restates them, though it adds real value by stating the `thread`/`all` mutual exclusivity and the origin of `thread` values from send_request. That is marginal added meaning above a fully documented schema.

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 names the exact resource ('The Agent's unacknowledged events, oldest first') and enumerates the event kinds returned (reply/in_reply_to, initiate, invitation, paired, receipt), so an agent knows precisely what comes back. It is clearly distinct from sibling tools such as acknowledge and send_request.

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?

It gives explicit routing rules: use `thread` 'whenever you're following up a request', use `all: true` only 'when the person asks for all of the Agent's replies, or to see who paired or started a request', and enable receipts only to check delivery/open status. It also states the follow-up action ('call acknowledge once handled'), leaving no inference needed.

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

All Taskriz tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?