Lakera

aid: lakera
url: https://raw.githubusercontent.com/api-evangelist/lakera/refs/heads/main/apis.yml
name: Lakera
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
- AI Security
- LLM Security
- Prompt Injection
- Guardrails
- Red Teaming
- GenAI
- API
description: Lakera is an AI security company building runtime defenses for generative AI applications. Its flagship Lakera
  Guard API screens prompts and responses for prompt injection, jailbreaks, PII leakage, unsafe content, and policy violations,
  while Lakera Red provides automated red teaming and risk assessment for GenAI systems. Lakera follows an API-first architecture
  with a managed SaaS platform, regional endpoints, self-hosted deployments for regulated environments, and a free Gandalf
  training game that has driven much of the prompt injection research community.
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
apis:
- aid: lakera:lakera-guard
  name: Lakera Guard
  tags:
  - AI Security
  - Prompt Injection
  - PII Detection
  - Content Moderation
  - LLM Guardrails
  humanURL: https://docs.lakera.ai/docs/api
  properties:
  - url: https://docs.lakera.ai/docs/api
    type: APIReference
  - url: https://docs.lakera.ai/docs/quickstart
    type: GettingStarted
  - url: https://platform.lakera.ai/docs/changelog
    type: ChangeLog
  description: Lakera Guard is a low-latency screening API that inspects text content sent to or from LLMs and flags threats
    including prompt injection, jailbreaks, PII, profanity, and policy violations. The /guard endpoint returns a flagged
    decision plus optional detector breakdowns via /guard/results, and the platform exposes /policies and /projects endpoints
    for managing detection logic. Available as a managed SaaS at api.lakera.ai/v2 with regional endpoints (US, EU, APAC)
    and as a self-hosted deployment for regulated environments.
- aid: lakera:lakera-red
  name: Lakera Red
  tags:
  - Red Teaming
  - AI Security
  - Vulnerability Assessment
  - GenAI Testing
  humanURL: https://www.lakera.ai/ai-red-teaming
  properties:
  - url: https://www.lakera.ai/ai-red-teaming
    type: Documentation
  description: Lakera Red is an automated red teaming product that probes GenAI applications for jailbreaks, prompt injection,
    data leakage, and other adversarial failures, then produces a risk-based remediation report. Designed for security teams
    evaluating AI applications prior to production deployment and for ongoing assurance of agentic systems.
- aid: lakera:gandalf
  name: Gandalf
  tags:
  - Education
  - Prompt Injection
  - Training
  - Research
  humanURL: https://gandalf.lakera.ai/
  properties:
  - url: https://gandalf.lakera.ai/
    type: Documentation
  description: Gandalf is Lakera's free interactive game that teaches prompt injection by challenging players to extract
    a secret from an LLM across progressively harder defenses. It has become a primary research and training environment
    for the prompt injection community and a public funnel into the Lakera Guard ecosystem.
common:
- type: Website
  url: https://www.lakera.ai/
- type: Documentation
  url: https://docs.lakera.ai/
- type: APIReference
  url: https://docs.lakera.ai/docs/api
- type: GettingStarted
  url: https://docs.lakera.ai/docs/quickstart
- type: ChangeLog
  url: https://platform.lakera.ai/docs/changelog
- type: Blog
  url: https://www.lakera.ai/blog
- type: Pricing
  url: https://www.lakera.ai/pricing
- type: Login
  url: https://platform.lakera.ai/login
- type: SignUp
  url: https://platform.lakera.ai/
- type: GitHubOrganization
  url: https://github.com/lakeraai
- type: LinkedIn
  url: https://www.linkedin.com/company/lakeraai/
- type: Events
  url: https://www.lakera.ai/events
- type: Game
  url: https://gandalf.lakera.ai/
- type: Features
  data:
  - name: Prompt Injection Detection
    description: Detects direct and indirect prompt injection attempts targeting LLM-powered applications and agents.
  - name: Jailbreak Detection
    description: Identifies attempts to bypass system prompts, safety policies, and model guardrails.
  - name: PII Detection
    description: Screens prompts and responses for personally identifiable information leakage.
  - name: Content Moderation
    description: Flags unsafe, toxic, or policy-violating content in user inputs and model outputs.
  - name: Custom Policies
    description: Configurable detection policies scoped to projects, models, and use cases.
  - name: Regional Endpoints
    description: Hosted endpoints in US, EU, and APAC regions for data residency and latency.
  - name: Self-Hosted Deployment
    description: On-premises and VPC deployments for regulated industries that cannot send data to a SaaS.
  - name: MCP Server
    description: Model Context Protocol server for embedding Lakera Guard into AI clients like Claude Code and Cursor.
- type: UseCases
  data:
  - name: LLM Application Guardrails
    description: Screen inputs and outputs of chatbots, copilots, and RAG applications for prompt injection and unsafe content.
  - name: AI Agent Protection
    description: Inspect tool inputs, retrieved context, and agent reasoning steps for indirect prompt injection.
  - name: Workforce AI Security
    description: Govern employee use of public LLMs and prevent sensitive data exfiltration.
  - name: GenAI Red Teaming
    description: Run automated adversarial assessments against pre-production GenAI applications.
  - name: Regulated Industry Deployments
    description: Self-host Guard in financial services, healthcare, and government environments with strict data residency.
- type: Integrations
  data:
  - name: OpenAI
    description: Wrap OpenAI calls with Guard screening for input and output safety.
  - name: Anthropic
    description: Screen prompts and responses from Anthropic Claude models.
  - name: LangChain
    description: Drop-in callbacks and chains for integrating Guard into LangChain applications.
  - name: LlamaIndex
    description: Screen queries and retrieved context in LlamaIndex RAG pipelines.
  - name: Model Context Protocol
    description: MCP server for surfacing Lakera Guard inside Claude Code, Cursor, and other MCP clients.
  - name: Kubernetes
    description: Self-hosted Helm-based deployments with health, readiness, and liveness probes.
maintainers:
- FN: Kin Lane
  email: [email protected]