API Evangelist conversational phrasing for Gitdealflow Com Answer API

# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Gitdealflow Com Answer API
  version: 1.0.0
extends: openapi/gitdealflow-com-answer-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: 6
- target: $.paths['/api/answer'].get
  update:
    x-apievangelist-phrasing:
      intent: Get the single best answer to a question
      effect: read
      questions:
      - How can I get one citation-ready answer to a question about startup signals?
      - Does the answer come with a confidence score and a canonical URL I can verify?
      - What's the single best match GitDealFlow has for my question?
      instructions:
      - text: Get the single best answer to {q} with its supporting facts.
        slots:
          q: query.q
      - text: Answer {q} and include the canonical URL for verification.
        slots:
          q: query.q
      method: generated
      generated: '2026-10-01'
- target: $.paths['/api/ask'].get
  update:
    x-apievangelist-phrasing:
      intent: Get ranked candidate answers to a question
      effect: read
      questions:
      - Can I see several ranked candidate answers instead of just one?
      - How many candidate matches can I ask for when comparing answers?
      instructions:
      - text: Show the top {limit} ranked candidate answers for {q}.
        slots:
          limit: query.limit
          q: query.q
      - text: Rank candidate answers to {q} for an autocomplete list.
        slots:
          q: query.q
      method: generated
      generated: '2026-10-01'
- target: $.paths['/api/diligence.json'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a public-source diligence dossier on a company
      effect: read
      questions:
      - Who acquired this company and which funds publicly backed it?
      - Can I get a cited diligence dossier with M&A history for a startup?
      - What happens if the company isn't in the tracked corpus?
      instructions:
      - text: Build the diligence dossier for {company}.
        slots:
          company: query.company
      - text: Show the acquirers and public backers of {company} with sources.
        slots:
          company: query.company
      - text: List which entities the diligence dossier can answer for.
      method: generated
      generated: '2026-10-01'
- target: $.paths['/api/answers.json'].get
  update:
    x-apievangelist-phrasing:
      intent: Download the full Q&A corpus
      effect: read
      questions:
      - Can I download every agent query and FAQ for offline embedding?
      - Where do I get the whole Q&A corpus to fine-tune a model?
      instructions:
      - text: Download the full Q&A corpus of agent queries and FAQs.
      - text: Dump the unversioned answers corpus for embedding.
      method: generated
      generated: '2026-10-01'
- target: $.paths['/api/v1/answers.json'].get
  update:
    x-apievangelist-phrasing:
      intent: Download the Q&A corpus from the v1 path
      effect: read
      questions:
      - Is there a versioned v1 URL for the full answers corpus?
      - Can I pin my pipeline to the v1 path of the Q&A dump?
      instructions:
      - text: Fetch the full Q&A corpus from the versioned v1 path.
      - text: Pull the v1 alias of the answers corpus.
      method: generated
      generated: '2026-10-01'
- target: $.paths['/api/v1/faq.json'].get
  update:
    x-apievangelist-phrasing:
      intent: Get the curated standalone FAQ corpus
      effect: read
      questions:
      - Which curated FAQ entries have stable slugs that are safe to cite?
      - What's in the standalone FAQ set, separate from the agent-query firehose?
      instructions:
      - text: Get the curated standalone FAQ entries.
      - text: List the FAQ answers with their stable slugs.
      method: generated
      generated: '2026-10-01'