generated: '2026-09-19'
method: searched
source: openapi/fetch-price-com-openapi.yml, curated against https://fetch-price.com/docs/, the provider SKILL.md
  and mcp/fetch_price_mcp.py
description: Recommended x-agentic-access execution contracts, classified heuristically from the OpenAPI. A governance
  starting point for exposing this API to AI agents — review and bind audience per deployment. See research/curity/agentic-governance/.
summary:
  operations: 4
  by_action_class:
    acting: 1
    connected: 3
  by_consequence:
    write: 1
    read: 3
  human_in_the_loop_required: 0
operations:
- path: /api/query
  method: post
  operationId: queryProducts
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: required
    audience: null
    token:
      max-ttl: 3600
    escalation:
      human-in-the-loop: none
      triggers: []
    audit: recommended
    note: Spends one lookup of the plan quota per call; the only side effect. Results may contain affiliate-tracked
      URLs the provider asks the agent to disclose to the human.
- path: /api/agents/register
  method: post
  operationId: registerAgent
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /health
  method: get
  operationId: getHealth
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
- path: /api/stats
  method: get
  operationId: getStats
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
curation_note: 'Heuristic output reclassified queryProducts: it is a POST but semantically a read - a marketplace
  search that persists nothing for the caller, places no order and moves no money (the buy URL is handed to a human).
  The provider describes both agent-facing skills as search and a "read-only health check". registerAgent stays
  acting/write (creates an account and a key). audience left null to bind per deployment.'