Standard Privacy ROPA — SDK Reference

Version: v1 | Base URL: https://api.standard.dev/api/v1 Auth: Bearer token (header Authorization: Bearer <token>) Tenant: Required (header x-tenant-id: <uuid>)

Quick Start — Text to ROPA
Complete Lifecycle
Endpoints Reference
SDK TypeScript Client
Data Model
Compliance Philosophy

Quick Start — Text to ROPA {#quick-start}

The fastest way to create a ROPA entry is to send natural language text:

curl -X POST https://api.standard.dev/api/v1/privacy/processing-activities/from-text \
  -H "Authorization: Bearer $TOKEN" \
  -H "x-tenant-id: $TENANT_ID" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Nós coletamos emails, nomes e CPFs dos nossos clientes para enviar campanhas de marketing. Compartilhamos os dados com a Mailchimp para disparo de emails. Guardamos por 2 anos. Temos consentimento via formulário web."
  }'

Response (201 Created):

{
  "data": {
    "activity": {
      "id": "a1b2c3d4-...",
      "name": "Atividade: enviar campanhas de marketing",
      "status": "draft",
      "purpose": "enviar campanhas de marketing",
      "legal_basis_lgpd": "consent",
      "retention_period": "2 anos",
      "third_party_sharing": true
    },
    "field_reviews_created": 8,
    "pending_questions": [
      "Quais medidas de segurança são aplicadas?"
    ],
    "warnings": [
      "Dados sensíveis detectados. DPIA provavelmente necessário."
    ],
    "confidence": 75,
    "agent_model": "rule-based-v1",
    "compliance_assertion": false
  },
  "trace_id": "tr-xyz..."
}

⚠️ compliance_assertion is ALWAYS false. The system never certifies compliance. All values are suggestions requiring human review.

Complete Lifecycle {#complete-lifecycle}

1. POST /from-text           → Create activity from natural language
   or POST /                 → Create activity from structured JSON

2. GET /:id/completeness     → Check what's missing

3. POST /:id/data-subjects   → Add who is affected
4. POST /:id/data-categories → Add what data is collected
5. POST /:id/third-parties   → Add who receives data

6. POST /:id/screen          → Run DPIA/LIA/TIA screening

7. GET /:id/field-reviews    → See AI suggestions
8. PUT /:id/field-reviews/:r → Approve or reject each

9. POST /:id/status          → Transition to "under_review"

10. GET /:id/report           → Generate ROPA report (JSON or Markdown)

Status Flow

draft → needs_information → under_review → approved
  │                                          │
  └── archived                          rejected

Endpoints Reference {#endpoints-reference}

Activity CRUD

Method	Path	Description
`POST`	`/privacy/processing-activities`	Create activity
`GET`	`/privacy/processing-activities`	List activities (query: `?status=`, `?limit=`, `?offset=`)
`GET`	`/privacy/processing-activities/:id`	Get activity by ID
`PUT`	`/privacy/processing-activities/:id`	Update activity fields
`DELETE`	`/privacy/processing-activities/:id`	Soft-delete activity

Status & Completeness

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/status`	Transition status
`GET`	`/privacy/processing-activities/:id/completeness`	Get completeness analysis

Data Subjects

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/data-subjects`	Add data subjects (array)
`GET`	`/privacy/processing-activities/:id/data-subjects`	List data subjects
`DELETE`	`/privacy/processing-activities/:id/data-subjects/:subjectId`	Remove a data subject

Create body (array):

[
  { "category": "customers", "estimated_count": "~50000", "vulnerable_group": false },
  { "category": "employees", "estimated_count": "~200" }
]

Categories: employees, customers, prospects, partners, suppliers, minors, patients, students, citizens, visitors, contractors, other

Data Categories

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/data-categories`	Add data categories (array)
`GET`	`/privacy/processing-activities/:id/data-categories`	List data categories
`DELETE`	`/privacy/processing-activities/:id/data-categories/:categoryId`	Remove a data category

Create body (array):

[
  { "category_name": "Email addresses", "sensitivity": "personal" },
  { "category_name": "CPF (Tax ID)", "sensitivity": "sensitive", "source_of_data": "Web form" }
]

Sensitivity levels: personal, sensitive, anonymized, pseudonymized, children, financial, health, biometric, genetic, political, religious, sexual, criminal, other

Third Parties

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/third-parties`	Add third parties (array)
`GET`	`/privacy/processing-activities/:id/third-parties`	List third parties
`DELETE`	`/privacy/processing-activities/:id/third-parties/:partyId`	Remove a third party

Create body (array):

[
  {
    "name": "Mailchimp",
    "role": "processor",
    "country": "US",
    "purpose": "Email delivery",
    "data_shared": ["email", "name"],
    "transfer_mechanism": "standard_contractual_clauses"
  }
]

Roles: processor, controller, joint_controller, sub_processor, recipient, other

Transfer mechanisms: adequacy_decision, standard_contractual_clauses, binding_corporate_rules, consent, contractual_necessity, legal_obligation, public_interest, vital_interests, not_applicable, other

Screenings (DPIA / LIA / TIA)

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/screen`	Run automated screenings
`GET`	`/privacy/processing-activities/:id/screenings`	List previous screenings

Trigger rules:

Screening	Triggers When
DPIA	`large_scope_processing`, `systematic_monitoring`, `vulnerable_subjects`, or `automated_decision_making` is `true`
LIA	`legal_basis_lgpd` is `"legitimate_interest"`
TIA	`international_transfer` is `true`

Field Reviews (Human-in-the-Loop)

Method	Path	Description
`POST`	`/privacy/processing-activities/:id/field-reviews`	Create a field review
`GET`	`/privacy/processing-activities/:id/field-reviews`	List all field reviews
`PUT`	`/privacy/processing-activities/:id/field-reviews/:reviewId`	Approve/reject a review

Create body:

{
  "field_name": "purpose",
  "suggested_value": "Customer analytics",
  "source": "ai_suggestion",
  "comment": "Extracted from uploaded policy document"
}

Update body (approve/reject):

{
  "review_status": "approved",
  "comment": "Confirmed by DPO on 2026-05-11"
}

Statuses: pending, approved, rejected, needs_revision Sources: human, ai_suggestion, system_rule, import

AI Extraction (Text → ROPA)

Method	Path	Description
`POST`	`/privacy/processing-activities/from-text`	Extract ROPA from natural language

Body:

{
  "text": "Full description of the processing activity in Portuguese or English..."
}

The AI extracts: purpose, legal basis, retention period, data subjects, data categories, third parties, risk flags. Every extracted field becomes a field_review requiring human approval.

Reports

Method	Path	Description
`GET`	`/privacy/processing-activities/:id/report`	Generate ROPA report
	`?format=json`	JSON format (default)
	`?format=markdown`	Markdown with evidence matrix

Report includes:

Completeness score + can_submit status
Every field with origin tracking (declared, ai_suggested, ai_approved, system_inferred, pending)
Executive summary
Third party table
Screening results
Gap list + warnings
compliance_assertion: false (always)

SDK TypeScript Client {#sdk-typescript-client}

import type {
  CreatePrivacyActivityRequest,
  CreatePrivacyDataSubjectRequest,
  CreatePrivacyDataCategoryRequest,
  CreatePrivacyThirdPartyRequest,
  PrivacyActivityResponse,
  PrivacyThirdPartyResponse,
  PrivacyScreeningResponse,
  PrivacyFieldReviewResponse,
} from "@standard/schemas";

class StandardPrivacyClient {
  constructor(
    private baseUrl: string,
    private token: string,
    private tenantId: string,
  ) {}

  private async request<T>(method: string, path: string, body?: unknown): Promise<T> {
    const res = await fetch(`${this.baseUrl}/api/v1/privacy${path}`, {
      method,
      headers: {
        "Authorization": `Bearer ${this.token}`,
        "x-tenant-id": this.tenantId,
        "Content-Type": "application/json",
      },
      body: body ? JSON.stringify(body) : undefined,
    });
    if (!res.ok) {
      const error = await res.json();
      throw new Error(`[${res.status}] ${error.message ?? JSON.stringify(error)}`);
    }
    const json = await res.json();
    return json.data;
  }

  // ─── Lifecycle ─────────────────────────────────────────────────

  /** Extract a ROPA entry from natural language */
  async extractFromText(text: string) {
    return this.request<{
      activity: PrivacyActivityResponse;
      field_reviews_created: number;
      pending_questions: string[];
      warnings: string[];
      confidence: number;
      compliance_assertion: false;
    }>("POST", "/processing-activities/from-text", { text });
  }

  /** Create a structured activity */
  async createActivity(data: CreatePrivacyActivityRequest) {
    return this.request<PrivacyActivityResponse>("POST", "/processing-activities", data);
  }

  /** Get activity by ID */
  async getActivity(id: string) {
    return this.request<PrivacyActivityResponse>("GET", `/processing-activities/${id}`);
  }

  /** Check completeness */
  async getCompleteness(activityId: string) {
    return this.request<{
      completeness_score: number;
      can_be_submitted_for_review: boolean;
      missing_required: string[];
      blocking_issues: Array<{ code: string; message: string; severity: string }>;
    }>("GET", `/processing-activities/${activityId}/completeness`);
  }

  // ─── Relations ─────────────────────────────────────────────────

  async addThirdParties(activityId: string, parties: CreatePrivacyThirdPartyRequest[]) {
    return this.request<PrivacyThirdPartyResponse[]>("POST", `/processing-activities/${activityId}/third-parties`, parties);
  }

  // ─── Screening ─────────────────────────────────────────────────

  async runScreening(activityId: string) {
    return this.request<PrivacyScreeningResponse[]>("POST", `/processing-activities/${activityId}/screen`);
  }

  // ─── Field Reviews ─────────────────────────────────────────────

  async listFieldReviews(activityId: string) {
    return this.request<PrivacyFieldReviewResponse[]>("GET", `/processing-activities/${activityId}/field-reviews`);
  }

  async approveFieldReview(activityId: string, reviewId: string, comment?: string) {
    return this.request<PrivacyFieldReviewResponse>("PUT", `/processing-activities/${activityId}/field-reviews/${reviewId}`, {
      review_status: "approved",
      comment,
    });
  }

  async rejectFieldReview(activityId: string, reviewId: string, comment: string) {
    return this.request<PrivacyFieldReviewResponse>("PUT", `/processing-activities/${activityId}/field-reviews/${reviewId}`, {
      review_status: "rejected",
      comment,
    });
  }

  // ─── Report ────────────────────────────────────────────────────

  async generateReport(activityId: string, format: "json" | "markdown" = "json") {
    return this.request<Record<string, unknown>>("GET", `/processing-activities/${activityId}/report?format=${format}`);
  }
}

Usage Example

const client = new StandardPrivacyClient(
  "https://api.standard.dev",
  "eyJ...",
  "tenant-uuid"
);

// 1. Extract from text
const result = await client.extractFromText(
  "Coletamos emails e CPFs de clientes para marketing. " +
  "Compartilhamos com Mailchimp. Retemos por 2 anos. Temos consentimento."
);
console.log(`Activity ${result.activity.id} created with ${result.confidence}% confidence`);
console.log(`Pending questions: ${result.pending_questions.join(", ")}`);

// 2. Run screening
const screenings = await client.runScreening(result.activity.id);
const dpiaRequired = screenings.find(s => s.screening_type === "dpia")?.result === "required";
console.log(`DPIA required: ${dpiaRequired}`);

// 3. Review AI suggestions
const reviews = await client.listFieldReviews(result.activity.id);
for (const review of reviews) {
  if (review.source === "ai_suggestion" && review.review_status === "pending") {
    // DPO reviews each suggestion
    await client.approveFieldReview(result.activity.id, review.id, "Confirmed by DPO");
  }
}

// 4. Check completeness
const completeness = await client.getCompleteness(result.activity.id);
console.log(`Completeness: ${completeness.completeness_score}%`);

// 5. Generate report
const report = await client.generateReport(result.activity.id, "markdown");
console.log(report);

Data Model {#data-model}

erDiagram
    ACTIVITY ||--o{ DATA_SUBJECT : has
    ACTIVITY ||--o{ DATA_CATEGORY : has
    ACTIVITY ||--o{ THIRD_PARTY : shares_with
    ACTIVITY ||--o{ SCREENING : screened_by
    ACTIVITY ||--o{ FIELD_REVIEW : reviewed_by
    ACTIVITY ||--o{ SCF_CONTROL : mapped_to
    ACTIVITY ||--|| REPORT : generates

    ACTIVITY {
        uuid id PK
        uuid tenant_id FK
        string name
        string purpose
        enum legal_basis_lgpd
        string retention_period
        boolean third_party_sharing
        boolean international_transfer
        enum status
    }

    THIRD_PARTY {
        uuid id PK
        string name
        enum role
        string country
        enum transfer_mechanism
    }

    SCREENING {
        uuid id PK
        enum screening_type
        enum result
        string recommendation
    }

    FIELD_REVIEW {
        uuid id PK
        string field_name
        enum source
        enum review_status
        uuid reviewer_id
    }

Compliance Philosophy {#compliance-philosophy}

The system NEVER asserts compliance. compliance_assertion: false is hardcoded.
AI suggests, humans decide. Every AI-extracted field creates a field_review requiring human approval.
Critical legal fields get double review. legal_basis_lgpd, retention_period, purpose, and dpia_required always get a system_rule review in addition to the AI review.
Absence of evidence ≠ absence of implementation. The system reports gaps, not failures.
Tenant isolation is absolute. Every query is scoped by tenant_id. Cross-tenant access returns empty, never throws (preventing information leakage).

Standard Privacy ROPA — SDK Reference

Standard Privacy ROPA — SDK Reference

Table of Contents

Quick Start — Text to ROPA {#quick-start}

Complete Lifecycle {#complete-lifecycle}

Status Flow

Endpoints Reference {#endpoints-reference}

Activity CRUD

Status & Completeness

Data Subjects

Data Categories

Third Parties

Screenings (DPIA / LIA / TIA)

Field Reviews (Human-in-the-Loop)

AI Extraction (Text → ROPA)

Reports

SDK TypeScript Client {#sdk-typescript-client}

Usage Example

Data Model {#data-model}

Compliance Philosophy {#compliance-philosophy}