Pangea AI Guard API

Detect and redact malicious content in LLM inputs and outputs.

Documentation

Specifications

Other Resources

Work with this as data

Every API here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for apis

7 MCP tools reach this
  • find_apisBrowse and filter every API in the catalog.
  • get_api_artifactsOne API's artifacts, grouped by type.
  • get_openapiThe primary OpenAPI for this API.
  • find_similar_apisAPIs that look like this one.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools

Call it yourself

curl for this page
This API
curl "https://apis.io/api/v1/apis/pangea-ai-guard-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no email required.

A second provider on the same verified email joins the account you already have.

OpenAPI Specification

pangea-ai-guard-api-openapi.yml Raw ↑
openapi: 3.0.1
info:
  title: Pangea Security Services AI Guard API
  description: Specification of representative Pangea security service APIs. Pangea exposes each security capability as its own REST service reachable at https://{service}.{csp}.{geo}.pangea.cloud (for example https://redact.aws.us.pangea.cloud). All requests are POST with a JSON body and are authenticated with a Bearer service token (or OAuth 2 access token) in the Authorization header. This document models several representative services - AuthN, Secure Audit Log, Redact, Vault, File Scan, IP Intel, Domain Intel, and AI Guard - and is not exhaustive of every endpoint or field.
  termsOfService: https://pangea.cloud/legal/terms-of-service/
  contact:
    name: Pangea Support
    url: https://pangea.cloud/contact/
  version: '1.0'
servers:
- url: https://{service}.{csp}.{geo}.pangea.cloud
  description: Per-service Pangea host. Each service is reached at its own subdomain.
  variables:
    service:
      default: redact
      description: Service name (authn, audit, redact, vault, file-scan, ip-intel, domain-intel, ai-guard).
    csp:
      default: aws
      description: Cloud service provider hosting the service.
    geo:
      default: us
      description: Geographic region (us, eu).
security:
- bearerAuth: []
tags:
- name: AI Guard
  description: Detect and redact malicious content in LLM inputs and outputs.
paths:
  /v1/text/guard:
    servers:
    - url: https://ai-guard.{csp}.{geo}.pangea.cloud
      variables:
        csp:
          default: aws
        geo:
          default: us
    post:
      operationId: aiGuardText
      tags:
      - AI Guard
      summary: Guard LLM text.
      description: Detect, remove, or block malicious content and intent in LLM inputs and outputs to prevent model manipulation and data leakage.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/AiGuardRequest'
      responses:
        '200':
          description: Guard result.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/AiGuardResponse'
        '401':
          $ref: '#/components/responses/Unauthorized'
components:
  schemas:
    PangeaResponse:
      type: object
      description: Standard Pangea response envelope wrapping every service result.
      properties:
        request_id:
          type: string
        request_time:
          type: string
          format: date-time
        response_time:
          type: string
          format: date-time
        status:
          type: string
          example: Success
        summary:
          type: string
        result:
          type: object
    AiGuardResponse:
      allOf:
      - $ref: '#/components/schemas/PangeaResponse'
      - type: object
        properties:
          result:
            type: object
            properties:
              blocked:
                type: boolean
              transformed:
                type: boolean
              recipe:
                type: string
              detectors:
                type: object
                description: Per-detector analysis (prompt_injection, pii, secrets, malicious_entity).
              prompt_text:
                type: string
    AiGuardRequest:
      type: object
      properties:
        text:
          type: string
          description: Plain text input (up to 20 KiB).
        messages:
          type: array
          items:
            type: object
            properties:
              role:
                type: string
              content:
                type: string
        recipe:
          type: string
          default: pangea_prompt_guard
        debug:
          type: boolean
          default: false
  responses:
    Unauthorized:
      description: Missing or invalid authentication token.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/PangeaResponse'
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: Pangea service token or OAuth 2 access token passed as a Bearer token.