# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Harness Pipeline Execute API
  version: 1.0.0
extends: openapi/harness-pipeline-execute-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-09-26'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 9
- target: $.paths['/pipeline/api/pipeline/execute/retryHistory/{planExecutionId}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get the retry history of an execution
      effect: read
      questions:
      - How many times has a pipeline execution been retried?
      - Where can I see the earlier attempts of a retried run?
      instructions:
      - text: Show the retry history for execution {planExecutionId} of pipeline {pipelineIdentifier}.
        slots:
          planExecutionId: path.planExecutionId
          pipelineIdentifier: query.pipelineIdentifier
      - text: List previous retry attempts of run {planExecutionId}.
        slots:
          planExecutionId: path.planExecutionId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/interrupt/{planExecutionId}'].put
  update:
    x-apievangelist-phrasing:
      intent: Abort, pause or resume a pipeline run
      effect: destructive
      questions:
      - How do I abort a running pipeline execution?
      - Can I pause and later resume a whole pipeline run?
      instructions:
      - text: Send interrupt {interruptType} to pipeline execution {planExecutionId}.
        slots:
          interruptType: query.interruptType
          planExecutionId: path.planExecutionId
      - text: Abort the whole run {planExecutionId} with interrupt {interruptType}.
        slots:
          planExecutionId: path.planExecutionId
          interruptType: query.interruptType
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/manual-execution/{nodeExecutionId}'].post
  update:
    x-apievangelist-phrasing:
      intent: Fail or resume a manual intervention step
      effect: write
      questions:
      - What do I do when a step is waiting on manual intervention?
      - Can I mark a paused manual step as failed instead of resuming it?
      instructions:
      - text: Mark manual step {nodeExecutionId} as {action}.
        slots:
          nodeExecutionId: path.nodeExecutionId
          action: requestBody.action
      - text: Resume the manual intervention on node {nodeExecutionId} with action {action}.
        slots:
          nodeExecutionId: path.nodeExecutionId
          action: requestBody.action
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/interrupt/{planExecutionId}/{nodeExecutionId}'].put
  update:
    x-apievangelist-phrasing:
      intent: Interrupt a single stage of a run
      effect: destructive
      questions:
      - Can I abort just one stage without stopping the rest of the pipeline?
      - Which interrupt types can be applied to an individual stage?
      instructions:
      - text: Send {interruptType} to stage {nodeExecutionId} in run {planExecutionId}.
        slots:
          interruptType: query.interruptType
          nodeExecutionId: path.nodeExecutionId
          planExecutionId: path.planExecutionId
      - text: Abort only stage node {nodeExecutionId} of execution {planExecutionId} using {interruptType}.
        slots:
          nodeExecutionId: path.nodeExecutionId
          planExecutionId: path.planExecutionId
          interruptType: query.interruptType
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/retry/{identifier}'].post
  update:
    x-apievangelist-phrasing:
      intent: Retry a failed pipeline run
      effect: write
      questions:
      - How do I rerun a failed pipeline from the stages that failed?
      - Can I retry every stage instead of just the failed ones?
      instructions:
      - text: Retry pipeline {identifier} execution {planExecutionId} from stages {retryStages}.
        slots:
          identifier: path.identifier
          planExecutionId: query.planExecutionId
          retryStages: query.retryStages
      - text: Rerun execution {planExecutionId} of {identifier} starting at {retryStages}, with note {notesForPipelineExecution}.
        slots:
          planExecutionId: query.planExecutionId
          identifier: path.identifier
          retryStages: query.retryStages
          notesForPipelineExecution: query.notesForPipelineExecution
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/retry/v2/{identifier}'].post
  update:
    x-apievangelist-phrasing:
      intent: Retry a run with new runtime inputs (V2)
      effect: write
      questions:
      - Can I change the runtime inputs when retrying a failed execution?
      - Is there a retry that also takes expression values?
      instructions:
      - text: Retry run {planExecutionId} of pipeline {identifier} at stages {retryStages} with runtime YAML {runtimeInputYaml}.
        slots:
          planExecutionId: query.planExecutionId
          identifier: path.identifier
          retryStages: query.retryStages
          runtimeInputYaml: requestBody.runtimeInputYaml
      - text: Use the V2 retry on {identifier} execution {planExecutionId} from {retryStages}, passing expression values {expressionValues}.
        slots:
          identifier: path.identifier
          planExecutionId: query.planExecutionId
          retryStages: query.retryStages
          expressionValues: requestBody.expressionValues
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/{identifier}/inputSetList'].post
  update:
    x-apievangelist-phrasing:
      intent: Run a pipeline using saved input sets
      effect: write
      questions:
      - Can I start a pipeline by referencing input sets I already saved?
      - What happens when several saved input sets are passed to one run?
      instructions:
      - text: Run pipeline {identifier} with input sets {inputSetReferences}.
        slots:
          identifier: path.identifier
          inputSetReferences: requestBody.inputSetReferences
      - text: Execute {identifier} on branch {branch} using saved input sets {inputSetReferences}.
        slots:
          identifier: path.identifier
          branch: query.branch
          inputSetReferences: requestBody.inputSetReferences
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/{identifier}'].post
  update:
    x-apievangelist-phrasing:
      intent: Run a pipeline with runtime input YAML
      effect: write
      questions:
      - How do I trigger a pipeline run through the Harness API?
      - Can I add a note to a pipeline run when I start it?
      instructions:
      - text: Run pipeline {identifier} in project {projectIdentifier}.
        slots:
          identifier: path.identifier
          projectIdentifier: query.projectIdentifier
      - text: Execute {identifier} from branch {branch} with note {notesForPipelineExecution}.
        slots:
          identifier: path.identifier
          branch: query.branch
          notesForPipelineExecution: query.notesForPipelineExecution
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pipeline/api/pipeline/execute/{identifier}/stages'].post
  update:
    x-apievangelist-phrasing:
      intent: Run selected stages of a pipeline
      effect: write
      questions:
      - Can I run only some stages of a pipeline instead of the whole thing?
      - Is it possible to pass runtime inputs when running selected stages?
      instructions:
      - text: Run only stages {stageIdentifiers} of pipeline {identifier}.
        slots:
          stageIdentifiers: requestBody.stageIdentifiers
          identifier: path.identifier
      - text: Execute stages {stageIdentifiers} of {identifier} with runtime inputs {runtimeInputYaml}.
        slots:
          stageIdentifiers: requestBody.stageIdentifiers
          identifier: path.identifier
          runtimeInputYaml: requestBody.runtimeInputYaml
      method: generated
      generated: '2026-09-26'