run-pipeline

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Run Pipeline

运行流水线

Execute and monitor Harness pipeline runs via MCP.

通过MCP执行并监控Harness流水线运行。

Instructions

操作说明

Step 1: Find the Pipeline

步骤1：查找流水线

Call MCP tool: harness_search
Parameters:
  query: "<user's pipeline name or keyword>"
  resource_types: ["pipeline"]
  compact: true

Or list all pipelines in a project:

Call MCP tool: harness_list
Parameters:
  resource_type: "pipeline"
  org_id: "<organization>"
  project_id: "<project>"
  search_term: "<optional filter>"

Call MCP tool: harness_search
Parameters:
  query: "<user's pipeline name or keyword>"
  resource_types: ["pipeline"]
  compact: true

或列出项目中的所有流水线：

Call MCP tool: harness_list
Parameters:
  resource_type: "pipeline"
  org_id: "<organization>"
  project_id: "<project>"
  search_term: "<optional filter>"

Step 2: Get Pipeline Details

步骤2：获取流水线详情

Call MCP tool: harness_get
Parameters:
  resource_type: "pipeline"
  resource_id: "<pipeline_identifier>"
  org_id: "<organization>"
  project_id: "<project>"

Extract from response:

Required runtime inputs (
```
<+input>
```
fields)
Pipeline variables with types and defaults
Stage structure and deployment targets

Call MCP tool: harness_get
Parameters:
  resource_type: "pipeline"
  resource_id: "<pipeline_identifier>"
  org_id: "<organization>"
  project_id: "<project>"

从响应中提取：

必填运行时输入（
```
<+input>
```
字段）
带类型和默认值的流水线变量
阶段结构和部署目标

Step 3: Check Available Input Sets

步骤3：查看可用输入集

Call MCP tool: harness_list
Parameters:
  resource_type: "input_set"
  org_id: "<organization>"
  project_id: "<project>"

Input sets provide pre-configured values that can be used or overridden.

Call MCP tool: harness_list
Parameters:
  resource_type: "input_set"
  org_id: "<organization>"
  project_id: "<project>"

输入集提供可使用或覆盖的预配置值。

Step 4: Execute the Pipeline

步骤4：执行流水线

Call MCP tool: harness_execute
Parameters:
  resource_type: "pipeline"
  action: "run"
  resource_id: "<pipeline_identifier>"
  org_id: "<organization>"
  project_id: "<project>"

Call MCP tool: harness_execute
Parameters:
  resource_type: "pipeline"
  action: "run"
  resource_id: "<pipeline_identifier>"
  org_id: "<organization>"
  project_id: "<project>"

Step 5: Monitor Execution

步骤5：监控执行状态

Call MCP tool: harness_get
Parameters:
  resource_type: "execution"
  resource_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"

Execution statuses: Running, Success, Failed, Aborted, Waiting (approval/input), Expired

Call MCP tool: harness_get
Parameters:
  resource_type: "execution"
  resource_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"

执行状态：Running、Success、Failed、Aborted、Waiting（等待审批/输入）、Expired

Step 6: Handle Approvals

步骤6：处理审批

If execution is waiting for approval:

Call MCP tool: harness_list
Parameters:
  resource_type: "approval_instance"
  org_id: "<organization>"
  project_id: "<project>"

Then approve or reject:

Call MCP tool: harness_execute
Parameters:
  resource_type: "approval_instance"
  action: "approve"  # or "reject"
  resource_id: "<approval_id>"

如果执行处于等待审批状态：

Call MCP tool: harness_list
Parameters:
  resource_type: "approval_instance"
  org_id: "<organization>"
  project_id: "<project>"

然后进行审批或拒绝：

Call MCP tool: harness_execute
Parameters:
  resource_type: "approval_instance"
  action: "approve"  # or "reject"
  resource_id: "<approval_id>"

Safety: Production Deployments

安全注意：生产环境部署

CRITICAL: Before executing pipelines targeting production:

Confirm intent with the user explicitly
Show what will be deployed (version, service, environment)
Check if pipeline has approval gates
Verify the version was tested in lower environments

重要提示：在执行面向生产环境的流水线之前：

明确与用户确认执行意图
展示将要部署的内容（版本、服务、环境）
检查流水线是否有审批关卡
验证该版本已在低级别环境中完成测试

Response Format

响应格式

Before Execution

执行前

Pipeline: <name>
Project: <project>
Required Inputs: <list variables needing values>
Available Input Sets: <list>
Ready to execute? Confirm to proceed.

Pipeline: <name>
Project: <project>
Required Inputs: <list variables needing values>
Available Input Sets: <list>
Ready to execute? Confirm to proceed.

After Trigger

触发后

Pipeline: <name>
Execution ID: <id>
Status: Running
View in Harness: <openInHarness link>

Pipeline: <name>
Execution ID: <id>
Status: Running
View in Harness: <openInHarness link>

Execution Complete

执行完成

Pipeline: <name>
Status: Success/Failed
Duration: <time>
Stages: <stage results table>

Pipeline: <name>
Status: Success/Failed
Duration: <time>
Stages: <stage results table>

Retrying Failed Executions

重试失败的执行

Retry is an action on

pipeline

(not

execution

). Pass the pipeline ID as

resource_id

and the failed execution's ID as

execution_id

Call MCP tool: harness_execute
Parameters:
  resource_type: "pipeline"
  action: "retry"
  resource_id: "<pipeline_id>"
  execution_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"

重试是针对

pipeline

的操作（而非

execution

）。将流水线ID作为

resource_id

传入，失败执行的ID作为

execution_id

传入：

Call MCP tool: harness_execute
Parameters:
  resource_type: "pipeline"
  action: "retry"
  resource_id: "<pipeline_id>"
  execution_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"

Interrupting Running Executions

中断运行中的执行

To abort an entire pipeline execution:

Call MCP tool: harness_execute
Parameters:
  resource_type: "execution"
  action: "interrupt"
  resource_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"
  params:
    interrupt_type: "AbortAll"

The

interrupt_type

field inside

params

is required. Valid values:

interrupt_type	Effect
`AbortAll`	Abort the entire pipeline execution
`UserMarkedFailure`	Mark the execution as failed by user

To abort multiple stuck executions, call

harness_execute

with

params: {interrupt_type: "AbortAll"}

for each execution ID.

要中止整个流水线执行：

Call MCP tool: harness_execute
Parameters:
  resource_type: "execution"
  action: "interrupt"
  resource_id: "<execution_id>"
  org_id: "<organization>"
  project_id: "<project>"
  params:
    interrupt_type: "AbortAll"

params

中的

interrupt_type

字段为必填项。有效值：

interrupt_type	作用
`AbortAll`	中止整个流水线执行
`UserMarkedFailure`	将执行标记为用户手动失败

要中止多个卡住的执行，需为每个执行ID调用

harness_execute

并传入

params: {interrupt_type: "AbortAll"}

。

Examples

示例

"Run the ci-pipeline" - Search for pipeline, execute with defaults
"Deploy version 2.0.0 to staging" - Find deploy pipeline, provide version input, execute
"What's the status of execution xyz123?" - Get execution details
"Retry the last failed deployment" - List recent failed executions, retry
"Abort all stuck executions" - List running/waiting executions, interrupt each with
```
params: {interrupt_type: "AbortAll"}
```
"Stop execution abc123" - Interrupt the specific execution with
```
params: {interrupt_type: "AbortAll"}
```

"Run the ci-pipeline" - 搜索流水线，使用默认值执行
"Deploy version 2.0.0 to staging" - 查找部署流水线，提供版本输入，执行
"What's the status of execution xyz123?" - 获取执行详情
"Retry the last failed deployment" - 列出近期失败的执行，进行重试
"Abort all stuck executions" - 列出运行中/等待中的执行，逐个调用
```
params: {interrupt_type: "AbortAll"}
```
进行中断
"Stop execution abc123" - 使用
```
params: {interrupt_type: "AbortAll"}
```
中断指定执行

Performance Notes

性能注意事项

Always confirm the pipeline identifier and required inputs before triggering execution. A failed execution due to missing inputs wastes time.
When monitoring executions, wait for terminal status (Success, Failed, Aborted) before reporting results. Do not assume intermediate states are final.
If an execution fails, gather full error context from logs before suggesting fixes. Do not guess at root causes.

触发执行前，务必确认流水线标识符和必填输入项。因缺少输入导致的执行失败会浪费时间。
监控执行状态时，需等待终端状态（Success、Failed、Aborted）后再报告结果。请勿假设中间状态为最终状态。
如果执行失败，需从日志中收集完整的错误上下文后再建议修复方案。请勿猜测根本原因。

Troubleshooting

故障排除

Pipeline Won't Start

流水线无法启动

Check delegate availability with
```
harness_status
```
Validate all required inputs have values
Test connector credentials with
```
harness_execute
```
(resource_type: "connector", action: "test_connection")

使用
```
harness_status
```
检查代理可用性
验证所有必填输入项均已赋值
使用
```
harness_execute
```
测试连接器凭据（resource_type: "connector", action: "test_connection"）

Missing Inputs

缺少输入项

Use
```
harness_get
```
on the pipeline to see all
```
<+input>
```
fields
Check input sets for pre-configured values

对流水线调用
```
harness_get
```
查看所有
```
<+input>
```
字段
检查输入集获取预配置值

Execution Stuck

执行卡住

Check for pending approvals with
```
harness_list
```
(resource_type: "approval_instance")
Check delegate status with
```
harness_status
```
Abort stuck executions with
```
harness_execute
```
(resource_type: "execution", action: "interrupt", params: {interrupt_type: "AbortAll"})

使用
```
harness_list
```
检查待处理审批（resource_type: "approval_instance"）
使用
```
harness_status
```
检查代理状态
使用
```
harness_execute
```
中止卡住的执行（resource_type: "execution", action: "interrupt", params: {interrupt_type: "AbortAll"}）