Skip to content
SPECIFICTION
Get listed

Pangea AI Guard

pangea.cloud

APISelf-hostedBrowser extension
Visit pangea.cloud
Spec score 7.4 #7 of 31 in AI Agent Security Software

What it is

Pangea AI Guard protects data and interactions in applications that use large language models. It detects prompt injection, jailbreak attempts, malicious entities, personal and confidential information, secrets, language, code, competitors, custom entities, and restricted topics. Early-access features address images and prompt hardening. Configurable recipes combine detectors and can report, block, redact, encrypt, or defang detected content at stages including prompts, RAG ingestion, model responses, and agent tool inputs or outputs. Detection uses heuristics, machine-learning classifiers, and fine-tuned LLMs. Applications connect through APIs, SDKs, or gateway integrations for Portkey, Kong, and LiteLLM. SDK documentation covers Python, Node.js, Go, Java, and C#. Deployment options are SaaS, Edge, and Private Cloud; Edge and Private Cloud keep processing within the customer’s cloud boundary. Enabling AI Guard activates an Activity Log for calls, inputs, outputs, detections, and context. The current request limit is 20 KiB of text.

Spec sheet

Purpose
AI Guard protects data and interactions with LLMs in AI-powered applications by safeguarding sensitive information, blocking malicious content, and recording application events.Read 3 Oct 2026 · pangea.cloud
Intended users
Pangea describes AI Guard as protection for AI-powered applications, including direct LLM interactions, agent data exchanges, and RAG data ingestion and retrieval.Read 3 Oct 2026 · pangea.cloud
Deployment
Pangea lists SaaS, Edge, and Private Cloud deployment models; Private Cloud runs data and control planes in the customer's infrastructure and is marked early access.Read 3 Oct 2026 · pangea.cloud
Integrations
AI Guard can be integrated through APIs and SDKs or through gateway extensions and plugins for Portkey, Kong, and LiteLLM.Read 3 Oct 2026 · pangea.cloud
Support
Pangea lists [email protected] for general inquiries and [email protected] for technical support.Read 3 Oct 2026 · pangea.cloud
Headquarters
636 Ramona St, Palo Alto, CA 94301Read 28 Sep 2026 · pangea.cloud
Company
Pangea’s company page lists its address as 636 Ramona St, Palo Alto, CA 94301.Read 2 Oct 2026 · pangea.cloud
Alerts
Users can configure webhooks and link them to specific detections to receive activity notifications and take action.Read 2 Oct 2026 · pangea.cloud
Audit logging
When AI Guard is enabled, its Activity Log is automatically activated and records service-call details including input, output, detections, and contextual information.Read 2 Oct 2026 · pangea.cloud
Audit trail
AI Guard automatically activates an Activity Log that records service calls, inputs, outputs, detections, and contextual information in the project audit trail.Read 3 Oct 2026 · pangea.cloud
Configurable actions
Recipes combine configurable detectors that can report, block, redact, encrypt, or defang detected content.Read 2 Oct 2026 · pangea.cloud
Detection
Its detectors cover prompt injection, malicious entities, PII and confidential data, secrets, language, code, competitors, custom entities, restricted topics, and early access image and prompt-hardening features.Read 2 Oct 2026 · pangea.cloud
Detection and sanitization
It detects prompt injection and other threats, filters content, and can redact or encrypt sensitive information in submitted text.Read 3 Oct 2026 · pangea.cloud
Detection methods
Threat and sensitive-data detection uses heuristics, machine-learning classifiers, and fine-tuned LLMs.Read 2 Oct 2026 · pangea.cloud
Detector coverage
Available detectors cover malicious prompts and entities, PII and confidential data, secrets, language, code, competitors, custom entities, restricted topics, and early-access image and prompt-hardening features.Read 3 Oct 2026 · pangea.cloud
Free account
Pangea’s getting-started guide says a free Pangea account is sufficient to enable AI Guard and start using its APIs.Read 2 Oct 2026 · pangea.cloud
Gateway integrations
Documented AI and API gateway integrations include Portkey, Kong, and LiteLLM.Read 2 Oct 2026 · pangea.cloud
Input limit
The current version supports processing up to 20 KiB of text.Read 2 Oct 2026 · pangea.cloud
Integration
Applications can connect through APIs or SDKs, and AI Guard documents SDKs for Python, JavaScript (Node.js), Go, Java, and C# through its integration options.Read 2 Oct 2026 · pangea.cloud
Pricing model
Pangea’s billing page lists AI Guard pricing details by data scanned, but the opened page showed only “Loading billing information...” and no price amount.Read 2 Oct 2026 · pangea.cloud
Recipes
Users can customize recipes made of detectors or create an unlimited number of recipes for stages such as prompts, RAG ingestion, LLM responses, and agent tool inputs and outputs.Read 3 Oct 2026 · pangea.cloud
Request limit
The current version supports processing up to 20 KiB of text per request.Read 3 Oct 2026 · pangea.cloud
Security certifications
The Pangea site displays SOC 2 Type 2, ISO/IEC 27001, and ISO/IEC 27701 compliance badges.Read 3 Oct 2026 · pangea.cloud
Supported gateway providers
The Kong plugins list Anthropic Claude, Azure OpenAI, Cohere, Google Gemini, Kong AI Gateway, and OpenAI as supported providers, and state that streaming responses are not currently supported.Read 3 Oct 2026 · pangea.cloud

Strong points and limits

Strong points

  • Detects prompt injection and jailbreak attempts
  • Recipes can redact, encrypt, or block content
  • Supports Python, Node.js, Go, Java, and C#
  • Edge and Private Cloud deployment options
  • Free plan available

Limits

  • Requests are limited to 20 KiB of text
  • Kong streaming responses are not currently supported

Who it suits

It suits teams building AI applications, agent exchanges, or RAG workflows that need configurable threat detection and audit logging. It is available through APIs, SDKs, extensions, or self-hosted deployment.

Verdict

Pangea AI Guard combines broad detector coverage with configurable remediation and automatic activity logging. Its free plan and multiple integration paths make it suitable for starting protection in an AI application, while the per-request text limit is an important constraint.

Where Specifiction places it

Alternatives to Pangea AI GuardAll alternatives →