Rein Agent Risk Scale

Returns the A-E Agent Risk Rating scale, what each grade means, the methodology version and the disclaimer.

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}

Score your own agent deployment from structured answers (enums and booleans). Returns an indicative grade (A-E, self-declared), the rated grade (E (unverified) until Rein verifies a KYB'd principal), the top 3 reasons and concrete steps that would raise the grade. Submissions are stored to build a pseudonymised cross-platform record: your answers, your grade, and -- as a label we never score -- the name your software gives itself (an MCP client name and version, or an SDK User-Agent). No IP address is stored. Your handle and that label are published only if you opt in; grades are never published without your agent's opt-in.

{
  "type": "object",
  "properties": {
    "principal": {
      "description": "Q1. Who is accountable for this agent? A registered business that has completed KYB (know-your-business) with a bank or payment provider; a registered business without KYB; a named individual; only a platform account claim (which does not count as an accountable principal); or no one.",
      "enum": [
        "registered_business_kyb_completed",
        "registered_business_no_kyb",
        "named_individual",
        "platform_claim_only",
        "none"
      ],
      "title": "Principal",
      "type": "string"
    },
    "repayment_path": {
      "description": "Q2. If you want credit, how would it be repaid? Only matters for the credit add-on.",
      "enum": [
        "not_seeking_credit",
        "receipts_into_account_lender_can_see_or_control",
        "third_party_revenue_with_evidence",
        "named_source_without_evidence",
        "none"
      ],
      "title": "Repayment Path",
      "type": "string"
    },
    "per_transaction_cap": {
      "description": "Q3a. Is there a per-transaction spend cap, and is it enforced outside the agent (server-side, by the card issuer or wallet) so the agent cannot raise or bypass it?",
      "enum": [
        "enforced_outside_agent",
        "in_agent_logic",
        "none"
      ],
      "title": "Per Transaction Cap",
      "type": "string"
    },
    "per_week_cap": {
      "description": "Q3b. Is there a per-week (or per-period) spend cap, and where is it enforced?",
      "enum": [
        "enforced_outside_agent",
        "in_agent_logic",
        "none"
      ],
      "title": "Per Week Cap",
      "type": "string"
    },
    "payee_binding": {
      "description": "Q4. Can the agent pay anyone, or only approved payees / a bound purpose?",
      "enum": [
        "allowlist_enforced_outside_agent",
        "purpose_bound_in_agent_logic",
        "none"
      ],
      "title": "Payee Binding",
      "type": "string"
    },
    "stop_switch": {
      "description": "Q5a. Is there a stop/kill switch? Fleet-level means it can halt every deployment sharing this model or harness, not just this one agent.",
      "enum": [
        "fleet_level_stop_with_named_holder",
        "operator_stop_outside_agent",
        "in_agent_only",
        "none"
      ],
      "title": "Stop Switch",
      "type": "string"
    },
    "stop_holder": {
      "description": "Q5b. Who holds the stop switch?",
      "enum": [
        "named_person_or_team_at_principal",
        "platform_or_infrastructure_provider",
        "independent_third_party",
        "the_agent_itself",
        "nobody"
      ],
      "title": "Stop Holder",
      "type": "string"
    },
    "human_escalation": {
      "description": "Q6. Does spend above a threshold require a human decision taken outside the agent?",
      "enum": [
        "required_above_threshold_enforced_outside_agent",
        "agent_decides_when_to_escalate",
        "none"
      ],
      "title": "Human Escalation",
      "type": "string"
    },
    "base_model_or_harness": {
      "default": "not_disclosed",
      "description": "Q7 (optional). Do you disclose the base model and harness? Disclosure is scored; which model you run is not.",
      "enum": [
        "disclosed_with_version",
        "disclosed_family_only",
        "not_disclosed"
      ],
      "title": "Base Model Or Harness",
      "type": "string"
    },
    "months_operating": {
      "description": "Q8a. How long has this agent been operating?",
      "enum": [
        "under_1",
        "1_to_3",
        "3_to_12",
        "over_12"
      ],
      "title": "Months Operating",
      "type": "string"
    },
    "payment_history": {
      "description": "Q8b. Payment history on any obligations so far.",
      "enum": [
        "none_yet",
        "some_all_on_time",
        "12_months_clean",
        "missed_or_late"
      ],
      "title": "Payment History",
      "type": "string"
    },
    "monitoring": {
      "description": "Q9. Is the agent's activity logged, and is anyone looking?",
      "enum": [
        "telemetry_shared_with_independent_party",
        "logged_and_reviewed_by_operator",
        "logged_not_reviewed",
        "none"
      ],
      "title": "Monitoring",
      "type": "string"
    },
    "incidents": {
      "description": "Q10. Any incidents (bad spend, off-purpose actions, compromise)?",
      "enum": [
        "none_known",
        "resolved_with_written_review",
        "unresolved",
        "prefer_not_to_say"
      ],
      "title": "Incidents",
      "type": "string"
    },
    "controls_evidence": {
      "default": "asserted",
      "description": "Q11 (optional). What backs your answers: your own assertion, an independent attestation, or an audit against Rein's factor definitions? Only an audit reaches A.",
      "enum": [
        "asserted",
        "attested_by_independent_party",
        "audited_against_rein_factors"
      ],
      "title": "Controls Evidence",
      "type": "string"
    },
    "agent_handle": {
      "anyOf": [
        {
          "pattern": "^[A-Za-z0-9_.\\-]{1,64}$",
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "description": "Optional public handle (e.g. your Moltbook name). Stored only if publish_opt_in is true. Letters, digits, _ . - only; max 64. Never scored.",
      "title": "Agent Handle"
    },
    "model_or_harness_name": {
      "anyOf": [
        {
          "maxLength": 80,
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "description": "Optional free text, max 80 chars. Stored for research; ignored by the scorer.",
      "title": "Model Or Harness Name"
    },
    "eval_session_id": {
      "anyOf": [
        {
          "pattern": "^[0-9a-f]{32}$",
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "description": "Optional. The session_id of a finished behaviour eval (eval mode). Its result is shown beside your indicative grade as evidence. Never scored: it does not change the grade.",
      "title": "Eval Session Id"
    },
    "publish_opt_in": {
      "default": false,
      "description": "Set true only if you consent to your grade being published with your handle. Defaults to false; nothing is published without it.",
      "title": "Publish Opt In",
      "type": "boolean"
    }
  },
  "required": [
    "principal",
    "repayment_path",
    "per_transaction_cap",
    "per_week_cap",
    "payee_binding",
    "stop_switch",
    "stop_holder",
    "human_escalation",
    "months_operating",
    "payment_history",
    "monitoring",
    "incidents"
  ],
  "additionalProperties": false,
  "description": "The ten questions (plus two optional labels and the opt-in flag)."
}

Explains what a grade on the Rein Agent Risk Scale requires and what moves a deployment up or down.

{
  "type": "object",
  "properties": {
    "grade": {
      "type": "string",
      "enum": [
        "A",
        "B",
        "C",
        "D",
        "E"
      ]
    }
  },
  "required": [
    "grade"
  ],
  "additionalProperties": false
}