API Evangelist conversational phrasing for Tooloracle Io Agent Safety API

# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Tooloracle Io Agent Safety API
  version: 1.0.0
extends: openapi/tooloracle-io-agent-safety-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: 5
- target: $.paths['/v2/agent_preflight'].post
  update:
    x-apievangelist-phrasing:
      intent: Pre-flight an agent action for GO or STOP
      effect: read
      questions:
      - Should my agent go ahead with this action, or is there a reason to stop?
      - Can I get a GO, CAUTION or STOP verdict with a cryptographic receipt before an agent acts?
      - What does the decision pre-flight check, such as evidence freshness and hallucination signals?
      instructions:
      - text: 'Pre-flight this agent intent and tell me GO, CAUTION or STOP: {intent}.'
        slots:
          intent: requestBody.intent
      - text: Check whether the agent should execute {intent} given session context {context}.
        slots:
          intent: requestBody.intent
          context: requestBody.context
      method: generated
      generated: '2026-10-01'
- target: $.paths['/v2/uvo_quick'].post
  update:
    x-apievangelist-phrasing:
      intent: Quickly gate an agent action proposal
      effect: read
      questions:
      - Is there a fast, fail-fast validation I can run on an agent's proposed action?
      - Which stages does the quick UVO action gate run before it returns APPROVE or BLOCK?
      instructions:
      - text: Run the quick UVO action gate on proposal {proposal}.
        slots:
          proposal: requestBody.proposal
      - text: Quick-check agent proposal {proposal} and stop at the first blocking stage.
        slots:
          proposal: requestBody.proposal
      method: generated
      generated: '2026-10-01'
- target: $.paths['/v2/uvo_citation_check'].post
  update:
    x-apievangelist-phrasing:
      intent: Verify legal and academic citations in a claim
      effect: read
      questions:
      - Can I catch a made-up MiCA or DORA article number before it goes into a report?
      - Which registers are citations checked against, like EUR-Lex and Semantic Scholar?
      - Will the citation check flag a fake DOI in a statement?
      instructions:
      - text: 'Verify the legal and academic citations in this claim: {claim}.'
        slots:
          claim: requestBody.claim
      - text: Check claim {claim} against sources {references} for fake citations.
        slots:
          claim: requestBody.claim
          references: requestBody.references
      method: generated
      generated: '2026-10-01'
- target: $.paths['/v2/uvo_full'].post
  update:
    x-apievangelist-phrasing:
      intent: Run the full UVO validation on an agent action
      effect: read
      questions:
      - What does the full UVO action gate return when I need the complete validation, not the quick one?
      - Can I get an APPROVE, WARN, ESCALATE or BLOCK decision with every stage's findings for an agent proposal?
      instructions:
      - text: Run the full UVO validation on agent proposal {proposal}.
        slots:
          proposal: requestBody.proposal
      - text: Give me the complete action-gate report, all stages, for {proposal}.
        slots:
          proposal: requestBody.proposal
      method: generated
      generated: '2026-10-01'
- target: $.paths['/v2/uvo_academic_verify'].post
  update:
    x-apievangelist-phrasing:
      intent: Deep-verify academic sources for a claim
      effect: read
      questions:
      - Can I check whether a research paper's journal is predatory?
      - Does the academic verifier resolve DOIs and cross-check the authors?
      instructions:
      - text: 'Deep-verify the academic sources behind this research statement: {claim}.'
        slots:
          claim: requestBody.claim
      - text: Resolve DOIs {references} and confirm authorship for claim {claim}.
        slots:
          claim: requestBody.claim
          references: requestBody.references
      method: generated
      generated: '2026-10-01'