# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Console Prompts API
  version: 1.0.0
extends: openapi/statsig-prompts-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-10-01'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 7
- target: $.paths['/console/v1/prompts'].get
  update:
    x-apievangelist-phrasing:
      intent: List prompts (AI Configs)
      effect: read
      questions:
      - What AI prompts are managed in my project?
      - Can I page through our AI Configs?
      instructions:
      - text: List all prompts.
      - text: Show prompts page {page}, {limit} per page.
        slots:
          page: query.page
          limit: query.limit
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts'].post
  update:
    x-apievangelist-phrasing:
      intent: Create a prompt (AI Config)
      effect: write
      questions:
      - How do I create a new managed prompt?
      - Can a new AI Config be scoped to target apps and a team?
      instructions:
      - text: Create a prompt named {name}.
        slots:
          name: requestBody.name
      - text: Create prompt {name} described as {description} for target apps {targetApps} owned by team {team}.
        slots:
          name: requestBody.name
          description: requestBody.description
          targetApps: requestBody.targetApps
          team: requestBody.team
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts/{id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a prompt
      effect: read
      questions:
      - What does a specific AI Config contain?
      - Can I read one prompt's settings?
      instructions:
      - text: Show prompt {id}.
        slots:
          id: path.id
      - text: Get the AI Config with ID {id}.
        slots:
          id: path.id
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts/{id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a prompt
      effect: destructive
      questions:
      - Can I delete a prompt we retired?
      - Is removing an AI Config permanent?
      instructions:
      - text: Delete prompt {id}.
        slots:
          id: path.id
      - text: Remove the AI Config {id}.
        slots:
          id: path.id
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts/{id}'].patch
  update:
    x-apievangelist-phrasing:
      intent: Update a prompt's description or ownership
      effect: write
      questions:
      - Can I change which team owns a prompt?
      - How do I update a prompt's target apps?
      instructions:
      - text: Move prompt {id} to team {team}.
        slots:
          id: path.id
          team: requestBody.team
      - text: Set prompt {id}'s description to {description} and target apps to {targetApps}.
        slots:
          id: path.id
          description: requestBody.description
          targetApps: requestBody.targetApps
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts/{id}/versions'].post
  update:
    x-apievangelist-phrasing:
      intent: Add a new version to a prompt
      effect: write
      questions:
      - How do I save a new version of a prompt with a different model?
      - Can a prompt version set temperature and max tokens?
      instructions:
      - text: Create version {name} of prompt {id} using model {model}.
        slots:
          id: path.id
          name: requestBody.name
          model: requestBody.model
      - text: Add version {name} to prompt {id} with messages {prompts}, temperature {temperature} and max tokens {max_tokens}.
        slots:
          id: path.id
          name: requestBody.name
          prompts: requestBody.prompts
          temperature: requestBody.temperature
          max_tokens: requestBody.max_tokens
      method: generated
      generated: '2026-10-01'
- target: $.paths['/console/v1/prompts/{id}/versions/{versionId}/start_evals'].post
  update:
    x-apievangelist-phrasing:
      intent: Start evals on a prompt version
      effect: write
      questions:
      - Can I kick off an evaluation run for one prompt version?
      - What's the way to test a new prompt version before rolling it out?
      instructions:
      - text: Start evals on version {versionId} of prompt {id}.
        slots:
          id: path.id
          versionId: path.versionId
      - text: Run the evaluation job for prompt {id} version {versionId}.
        slots:
          id: path.id
          versionId: path.versionId
      method: generated
      generated: '2026-10-01'