Skip to main content
Glama
DeepSeek Harness MCP Server

DeepSeek Harness MCP Server

dsh_get_task

Read-only

Retrieves task status, bounded assistant output, tool calls, and pending permissions; use the optional wait to poll for completion without blocking.

Instructions

Get task status, bounded assistant output, tool calls and pending permissions. wait_ms waits up to 10 seconds for completion. awaiting_permission needs dsh_respond_permission; do not keep polling without addressing it. failed/timed_out/cancelled may have partial file changes. completed means Harness ended its turn, not independent acceptance verification.

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": {
      "task_id": "<task_id>"
    },
    "name": "dsh_get_task"
  }
}

Input Schema

NameRequiredDescriptionDefault
task_idYes
wait_msNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.3/5.0
Behavior5/5

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

Annotations only declare readOnlyHint and openWorldHint=false; the description adds real behavioral context on top — blocking semantics of wait_ms, the polling anti-pattern around awaiting_permission, partial file changes on failed/timed_out/cancelled, and the caveat that 'completed' means the Harness ended its turn rather than verified acceptance. Those are exactly the semantics an agent cannot get from structured fields.

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

Conciseness5/5

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

Four dense sentences, each carrying distinct information: what is returned, how to wait, what to do on awaiting_permission, and status caveats. Front-loaded with the return contents and free of filler or restated annotation boilerplate.

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?

There is no output schema, so the description must cover return values and does — naming the four payload elements and qualifying assistant output as 'bounded'. Status semantics are covered for the risky states. It does not describe how much output is bounded or whether truncated output requires a follow-up call, which is the remaining gap.

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

Parameters4/5

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

Schema coverage is 0%, so the description carries the burden; it explains that wait_ms waits for completion up to 10 seconds, adding meaning beyond the bare integer/min/max, and task_id is self-evident as a UUID. The distinction between wait_ms=0 (non-blocking snapshot) and a positive value is only implied rather than stated.

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

Purpose4/5

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

States a specific verb ('Get') and enumerates exactly what is returned: task status, bounded assistant output, tool calls, and pending permissions. That is far more useful than a tautology, but it never contrasts itself with listed siblings such as dsh_list_tasks (which presumably also surfaces status), so an agent must infer the boundary.

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

Usage Guidelines4/5

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

Gives concrete branching guidance: when status is awaiting_permission, use dsh_respond_permission rather than re-polling, and wait_ms is the mechanism for blocking up to 10 seconds. It does not say when to prefer dsh_list_tasks or when to escalate to dsh_cancel_task, so it stops short of full alternative routing.

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