Skip to main content
Glama

Get conversation or run result

dsh_get
Read-onlyIdempotent

Collect the final result of a dispatched agent turn by conversation and run ID, including answer, tools, usage, errors and review ticket. Without a run ID, returns the most recent turn.

Instructions

Details of one conversation; with run_id, the final result of that dispatch (the place to collect asynchronous work). Without run_id it uses the most recent turn. Returns: title, workspace, permission tier, state, turn count, token usage, context pressure, pending approvals, and the turn's full result (answer, tools, usage, error) with its review ticket. The service keeps the latest 20 runs in memory. After a restart or eviction the result is read back from the sentinel file, which is on disk and survives restarts.

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": {
      "conversation_id": "<conversation_id>"
    },
    "name": "dsh_get"
  }
}

Input Schema

NameRequiredDescriptionDefault
run_idNoThe run_id returned by dsh_send. With only the conversation ID, look up the current or most recently started turn, including retained results after reconnecting.
max_charsNoResult budget (default 0 = no truncation, full result). Above 0 it applies to **both** text and structured output: the answer is cut to that length (with answer_truncated / next_offset), and structured.run.result.tools_used keeps only tool name, status and input_chars instead of full inputs. The complete result is always in sentinel_file. When a turn used many tools the default can exceed a client's per-result limit; about 8000 is a good value.
client_run_idNoLook the run up by its idempotency key instead of run_id. Works across restarts during sentinel retention.
conversation_idYesConversation id, from dsh_start or dsh_list.

Output Schema

NameRequiredDescriptionDefault
runYes
sentinel_fileNo
conversation_idYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.7.3
    • changedInput schema / properties / run_id / description
      Previous value: -"The run_id returned by dsh_send. Omit for the most recent turn."New value: +"The run_id returned by dsh_send. With only the conversation ID, look up the current or most recently started turn, including retained results after reconnecting."
  2. Changed2 schema fields changedv0.7.1
    • changedInput schema / properties / client_run_id / description
      Previous value: -"Look the run up by the caller's idempotency key instead of run_id (only works while the run is still in memory, i.e. within the latest 20)."New value: +"Look the run up by its idempotency key instead of run_id. Works across restarts during sentinel retention."
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "properties": {
      +    "conversation_id": {
      +      "type": "string"
      +    },
      +    "run": {
      +      "properties": {
      +        "result": {
      +          "type": [
      +            "object",
      +            "null"
      +          ]
      +        },
      +        "run_id": {
      +          "type": "string"
      +        },
      +        "status": {
      +          "type": "string"
      +        }
      +      },
      +      "required": [
      +        "run_id",
      +        "status",
      +        "result"
      +      ],
      +      "type": [
      +        "object",
      +        "null"
      +      ]
      +    },
      +    "sentinel_file": {
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "conversation_id",
      +    "run"
      +  ],
      +  "type": "object"
      +}
  3. Changed4 schema fields changedv0.6.8
    • changedInput schema / properties / client_run_id / description
      Previous value: -"按调用方幂等键取 run(与 run_id 二选一;只对仍在内存里的 run 有效,最多最近 20 条)。"New value: +"Look the run up by the caller's idempotency key instead of run_id (only works while the run is still in memory, i.e. within the latest 20)."
    • changedInput schema / properties / conversation_id / description
      Previous value: -"会话 id,来自 dsh_start 或 dsh_list。"New value: +"Conversation id, from dsh_start or dsh_list."
    • changedInput schema / properties / max_chars / description
      Previous value: -"结果预算(默认 0 = 不截断,返回完整结果)。>0 时**同时**作用于 text 与 structured:正文截到该长度(附 answer_truncated/next_offset),structured.run.result.tools_used 只保留工具名/状态与 input_chars,不再带完整输入。完整结果始终在 sentinel_file 里。一个回合用了很多工具时,默认返回可能超过客户端单次工具结果上限,建议传 8000 左右。"New value: +"Result budget (default 0 = no truncation, full result). Above 0 it applies to **both** text and structured output: the answer is cut to that length (with answer_truncated / next_offset), and structured.run.result.tools_used keeps only tool name, status and input_chars instead of full inputs. The complete result is always in sentinel_file. When a turn used many tools the default can exceed a client's per-result limit; about 8000 is a good value."
    • changedInput schema / properties / run_id / description
      Previous value: -"dsh_send 返回的 run_id;不传则取最近一次回合。"New value: +"The run_id returned by dsh_send. Omit for the most recent turn."
  4. Changed1 schema field changedv0.5.2
    • changedInput schema / properties / max_chars / description
      Previous value: -"正文在 **text** 里的最大字符数(默认 0 = 不截断)。structured.run.result 仍保留完整内容;截断时附带 answer_truncated/next_offset,可自行分页。"New value: +"结果预算(默认 0 = 不截断,返回完整结果)。>0 时**同时**作用于 text 与 structured:正文截到该长度(附 answer_truncated/next_offset),structured.run.result.tools_used 只保留工具名/状态与 input_chars,不再带完整输入。完整结果始终在 sentinel_file 里。一个回合用了很多工具时,默认返回可能超过客户端单次工具结果上限,建议传 8000 左右。"
  5. Changed3 schema fields changedv0.5.1
    • addedInput schema / properties / client_run_id
      Added value: +{
      +  "description": "按调用方幂等键取 run(与 run_id 二选一;只对仍在内存里的 run 有效,最多最近 20 条)。",
      +  "maxLength": 128,
      +  "type": "string"
      +}
    • addedInput schema / properties / max_chars
      Added value: +{
      +  "description": "正文在 **text** 里的最大字符数(默认 0 = 不截断)。structured.run.result 仍保留完整内容;截断时附带 answer_truncated/next_offset,可自行分页。",
      +  "type": "number"
      +}
    • addedInput schema / properties / run_id / maxLength
      Added value: +200
  6. First observedv0.1.0

TDQS

A4.2/5.0
Behavior3/5

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

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so the safety profile is covered. The description does add genuinely non-redundant behavior: the service holds only the latest 20 runs in memory, and after restart or eviction the result is re-read from a sentinel file on disk that survives restarts. That retention/eviction detail is useful but is the only behavioral disclosure beyond annotations, so 3 rather than higher.

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?

Front-loaded with the mode distinction, then returns, then retention. Every sentence carries information, but the long 'Returns:' field list partially duplicates the output schema and the bold emphasis adds noise. Slightly over-packed for the value delivered.

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?

With an output schema and full annotations present, the description only needs to cover call semantics, and it does: mode selection, the async-collection intent, truncation interplay with sentinel_file, and post-restart behavior. Nothing an agent needs in order to invoke it correctly is missing.

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 100%, so the baseline would be 3; the description earns above that by explaining the default path ('without run_id it uses the most recent turn'), which the schema does not state. It does not expand on max_chars or client_run_id semantics beyond what the schema already documents in detail, so it is not a 5.

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?

Names a specific verb and resource (details of one conversation / final result of a dispatch) and explicitly splits the two operating modes: with run_id returns the dispatch's final result, without it falls back to the most recent turn. That distinction alone lets an agent tell this apart from siblings like dsh_read or dsh_status without inspecting either schema.

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?

'The place to collect asynchronous work' gives a clear positive condition for calling it, and the run_id/no-run_id split tells the agent which mode to pick. It stops short of naming an alternative tool or an explicit when-not-to-use case, so it lands just below the top band.

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