generated: '2026-09-19'
method: searched
source: https://api.verse-me.com/.well-known/agent.json — AGP extension capabilitiesAnnounced[] policies + urn:verse:agent-extensions:v1 endpoint list + securityRequirements
description: >-
  Recommended x-agentic-access execution contracts for the twelve operations Verse's agent card names. There is no
  OpenAPI, so the heuristic derive-agentic-access.py could not run; the action-class and auth requirement are read
  from the provider's OWN Agent Gateway Protocol policy block (requiresAuth, cost, mode, dataScope per capability)
  and its securityRequirements (public vs x402Payment per skill), which is why the method is searched rather than
  generated. consequence is API Evangelist's reading: every paid call is an irreversible on-chain USDC transfer with
  no published refund path, so paid operations are rated physical with human-in-the-loop required and a token TTL
  ceiling of 300s; the free calibration/reputation reads are connected/read. TTL ceilings mirror the agentic-access
  Spectral ruleset. A governance starting point, not a provider claim; audience is left null to bind per deployment.
  See research/curity/agentic-governance/.
audience: null
custody_note: >-
  The caller pays from a wallet it controls; Verse never holds a caller credential. The consequence ratings describe
  what a signed x402 payment does once broadcast — it is final — not what Verse can do with the caller's keys.
summary:
  operations: 12
  by_action_class:
    connected: 6
    acting: 6
  by_consequence:
    read: 6
    physical: 6
  human_in_the_loop_required: 6
  provider_policy:
    capabilities_announced: 3
    requiresAuth_false: 2
    requiresAuth_true: 1
    payment_gated_skills: 3
    public_skills: 3
operations:
- operation: GET /expertise/calibration
  skill: calibration_track_record
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    provider-policy: {capability: 'expertise:calibration', requiresAuth: false, cost: 0, dataScope: aggregate_only}
- operation: GET /expertise/calibration/signed
  skill: signed_calibration
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    provider-policy: {capability: 'expertise:calibration:signed', requiresAuth: false, cost: 0, provenance: ed25519, dataScope: aggregate_only}
- operation: GET /expertise/calibration/history
  skill: calibration_history
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    provider-policy: {security: public}
- operation: GET /expertise/calibration/history/signed
  skill: calibration_history
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    provider-policy: {security: public, provenance: ed25519}
- operation: GET /reputation
  skill: null
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    provider-policy: {security: public (proof bundle)}
- operation: POST /expertise/consultation/price-estimate
  skill: consultation (preview)
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
    dry-run-for: POST /expertise/consultation
    provider-policy: {security: public, listed under freeEndpoints as "Preview cost before paying"}
- operation: POST /expertise/consultation
  skill: consultation
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: irreversible on-chain USDC payment ($3–$50 published); no refund or cancel path published; price is dynamic
    audit: required
    provider-policy: {capability: 'expertise:consultation', requiresAuth: true, cost: 0.1 (disagrees with the skill's $3–$50), mode: async, paymentProtocol: x402, paymentNetwork: 'eip155:8453', paymentAsset: USDC, provenance: ed25519}
- operation: POST /expertise/consultation (strategy_recommendation)
  skill: strategy_recommendation
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: paid via x402 with NO published price; the caller cannot bound the spend from the card
    audit: required
    provider-policy: {security: x402Payment}
- operation: POST /expertise/epistemic-audit
  skill: epistemic_audit
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: paid via x402 with NO published price; irreversible
    audit: required
    provider-policy: {security: x402Payment}
- operation: POST /expertise/edgar (index)
  skill: edgar_financial_intelligence
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: on-chain payment ($0.01 published); irreversible even at this size
    audit: required
    provider-policy: {security: x402Payment, price: '$0.01'}
- operation: POST /expertise/edgar (analysis)
  skill: edgar_financial_intelligence
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: on-chain payment ($0.05 published); irreversible
    audit: required
    provider-policy: {security: x402Payment, price: '$0.05'}
- operation: POST /expertise/edgar (deep)
  skill: edgar_financial_intelligence
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    purpose-required: true
    token:
      max-ttl: 300
      exchange: required
    escalation:
      human-in-the-loop: required
      reason: on-chain payment ($0.25 published); irreversible
    audit: required
    provider-policy: {security: x402Payment, price: '$0.25'}