# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Thehiveryiq Com Amplify API
  version: 1.0.0
extends: openapi/thehiveryiq-com-amplify-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['/v1/amplify/call'].post
  update:
    x-apievangelist-phrasing:
      intent: Run a prompt through the Amplify loop
      effect: write
      questions:
      - How do I send a prompt through Amplify so it's compressed, answered and certified?
      - Can I choose the model, max tokens and temperature for an Amplify call?
      instructions:
      - text: Run prompt {prompt} through Amplify for tenant {tenant_did}.
        slots:
          prompt: requestBody.prompt
          tenant_did: requestBody.tenant_did
      - text: Call Amplify with {prompt} on model {model} for tenant {tenant_did}, capped at {max_tokens} tokens.
        slots:
          prompt: requestBody.prompt
          model: requestBody.model
          tenant_did: requestBody.tenant_did
          max_tokens: requestBody.max_tokens
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/corpus/append'].post
  update:
    x-apievangelist-phrasing:
      intent: Add an entry to a tenant's Amplify corpus
      effect: write
      questions:
      - How can I add my own content to the corpus Amplify draws on?
      - What kinds of entries can I append to a tenant corpus?
      instructions:
      - text: Append {content} as a {kind} entry to tenant {tenant_did}'s corpus.
        slots:
          content: requestBody.content
          kind: requestBody.kind
          tenant_did: requestBody.tenant_did
      - text: 'Save this {kind} to the Amplify corpus for {tenant_did}: {content}.'
        slots:
          kind: requestBody.kind
          tenant_did: requestBody.tenant_did
          content: requestBody.content
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/corpus/retrieve'].get
  update:
    x-apievangelist-phrasing:
      intent: Search a tenant's Amplify corpus
      effect: read
      questions:
      - Can I pull the corpus entries most relevant to a query?
      - How many matches can a corpus retrieval return?
      instructions:
      - text: Search tenant {tenant_did}'s corpus for {query}.
        slots:
          tenant_did: query.tenant_did
          query: query.query
      - text: Retrieve the top {k} corpus entries for {query} in tenant {tenant_did}.
        slots:
          k: query.k
          query: query.query
          tenant_did: query.tenant_did
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/corpus/stats'].get
  update:
    x-apievangelist-phrasing:
      intent: Get Amplify corpus statistics
      effect: read
      questions:
      - How big is my tenant's Amplify corpus?
      - What stats does Amplify keep about the corpus?
      instructions:
      - text: Show corpus stats for tenant {tenant_did}.
        slots:
          tenant_did: query.tenant_did
      - text: Get overall Amplify corpus statistics.
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/sign'].post
  update:
    x-apievangelist-phrasing:
      intent: Certify an LLM response you ran yourself
      effect: write
      questions:
      - Can I get a signed certificate for an LLM answer without sharing my model API key?
      - What token savings would compression have given a prompt I already ran?
      instructions:
      - text: Certify response {response} to prompt {prompt} for tenant {tenant_did}.
        slots:
          response: requestBody.response
          prompt: requestBody.prompt
          tenant_did: requestBody.tenant_did
      - text: Sign my own {model} output {response} for prompt {prompt} under tenant {tenant_did}.
        slots:
          model: requestBody.model
          response: requestBody.response
          prompt: requestBody.prompt
          tenant_did: requestBody.tenant_did
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/verify'].post
  update:
    x-apievangelist-phrasing:
      intent: Verify an Amplify certificate
      effect: read
      questions:
      - How can anyone confirm an Amplify certificate's signature is valid?
      - Can I check that a certificate binds to a specific prompt and response?
      instructions:
      - text: Verify certificate {payload} with signature {signature_b64} and public key {pubkey_b64}.
        slots:
          payload: requestBody.payload
          signature_b64: requestBody.signature_b64
          pubkey_b64: requestBody.pubkey_b64
      - text: Check that certificate {payload} signed {signature_b64} under {pubkey_b64} matches response {response}.
        slots:
          payload: requestBody.payload
          signature_b64: requestBody.signature_b64
          pubkey_b64: requestBody.pubkey_b64
          response: requestBody.response
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/wrap-receipt'].post
  update:
    x-apievangelist-phrasing:
      intent: Upgrade a vertical receipt to a signed cert
      effect: write
      questions:
      - Can I turn an existing receipt from another vertical into a CCACW signed certificate?
      - Does wrapping a receipt change my original receipt?
      instructions:
      - text: Wrap receipt {receipt} from vertical {vertical} for event {event_type} with payload {payload}, tenant {tenant_did}.
        slots:
          receipt: requestBody.receipt
          vertical: requestBody.vertical
          event_type: requestBody.event_type
          payload: requestBody.payload
          tenant_did: requestBody.tenant_did
      - text: Add a CCACW cert to the {vertical} receipt {receipt} for {event_type}, payload {payload}, tenant {tenant_did}.
        slots:
          vertical: requestBody.vertical
          receipt: requestBody.receipt
          event_type: requestBody.event_type
          payload: requestBody.payload
          tenant_did: requestBody.tenant_did
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/selftest'].get
  update:
    x-apievangelist-phrasing:
      intent: Self-test the Amplify pipeline
      effect: read
      questions:
      - Is there a way to test Amplify's compile, compress and sign steps without calling an LLM?
      - What does the Amplify self-test verify?
      instructions:
      - text: Run the Amplify self-test.
      - text: Check the compile-compress-sign chain is working.
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/amplify/bench'].get
  update:
    x-apievangelist-phrasing:
      intent: Benchmark Amplify against a raw model call
      effect: read
      questions:
      - How much does Amplify reduce tokens, latency and cost versus a plain call?
      - Can I benchmark Amplify with my own prompt and model?
      instructions:
      - text: Benchmark Amplify.
      - text: Benchmark prompt {prompt} on model {model} with and without Amplify.
        slots:
          prompt: query.prompt
          model: query.model
      method: generated
      generated: '2026-09-26'