# PromptGuard

> AI security firewall. Protect LLM applications from prompt injection, data leaks, jailbreaks, and PII exposure.

## Docs

- [PromptGuard](https://docs.promptguard.co/index.md): AI security firewall between your app and any LLM provider. Scans every request and response for prompt injection, data leaks, jailbreaks, and PII exposure.
- [Why PromptGuard?](https://docs.promptguard.co/why-promptguard.md): What PromptGuard protects, the risks it addresses, and how to evaluate it — no code required
- [Quickstart](https://docs.promptguard.co/quickstart.md): Add PromptGuard to an existing LLM application in 5 minutes. Get an API key, install the SDK, and verify protection.
- [Security Overview](https://docs.promptguard.co/security/overview.md): Configure PromptGuard security policies to protect your AI applications
- [What Detection Can and Cannot Do](https://docs.promptguard.co/security/detection-limits.md): The measured ceiling on prompt-injection detection, why it exists, and which controls do not depend on it
- [Threat Detection](https://docs.promptguard.co/security/threat-detection.md): Advanced threat detection capabilities and configuration
- [Policy Presets](https://docs.promptguard.co/security/policy-presets.md): Composable security policy presets combining use-case templates with strictness levels
- [Custom Security Rules](https://docs.promptguard.co/security/custom-rules.md): Create granular policy rules to enforce domain-specific security requirements
- [Policy-as-Code](https://docs.promptguard.co/security/policy-as-code.md): Define guardrail configurations in YAML and manage them with git
- [MCP Server Security](https://docs.promptguard.co/security/mcp-security.md): Validate Model Context Protocol tool calls in agent architectures
- [Agent Trace Analysis](https://docs.promptguard.co/security/agent-traces.md): Submit a full agent run to catch dataflow and goal-alignment attacks no single message reveals
- [Capability Containment](https://docs.promptguard.co/security/capability-containment.md): Fail-closed policy derived from the user objective, so an injection cannot authorize what the user never asked for
- [Untrusted Content Marking](https://docs.promptguard.co/security/untrusted-content-marking.md): Mark retrieved and tool content so an instruction inside it cannot read as an instruction from you — plus the one model setting that outperforms it
- [Agent Memory Scanning](https://docs.promptguard.co/security/memory-scanning.md): Catch poisoned memory chunks before they persist — the injection that outlives the request that planted it
- [Gateway Overview](https://docs.promptguard.co/gateway/overview.md): How PromptGuard sits in front of your LLM provider and what it does on every request
- [Provider Keys](https://docs.promptguard.co/gateway/provider-keys.md): Store your OpenAI / Anthropic / Bedrock credentials once, then forward them server-side on every request
- [Allow & Block Lists](https://docs.promptguard.co/gateway/access-lists.md): Reject or whitelist requests by source IP, CIDR block, end-user ID, or country before they hit your policy engine
- [Routing Rules](https://docs.promptguard.co/gateway/routing-rules.md): Rewrite the upstream provider/model on the fly based on request shape, with optional smart failover
- [End-User Attribution](https://docs.promptguard.co/gateway/end-users.md): Track per-customer usage, cost, threats, and risk by passing a single header on every request
- [Shadow AI](https://docs.promptguard.co/shadow-ai/overview.md): Stop secrets and customer data leaking into ChatGPT, Claude, Gemini, and other AI tools your team uses — checked on the device, before anything is sent.
- [Get Started](https://docs.promptguard.co/shadow-ai/quickstart.md): Protect your team’s AI usage in a few clicks. No security expertise, no command line — open the app, sign in, done.
- [Test It Yourself](https://docs.promptguard.co/shadow-ai/test-drive.md): Install Shadow AI, turn it on, and try to leak a secret to Claude — the complete hands-on walkthrough, start to finish, about 15 minutes.
- [Desktop Agent](https://docs.promptguard.co/shadow-ai/desktop-agent.md): Install the Shadow AI agent and your Mac or PC checks every AI tool — ChatGPT, Claude, Cursor, and more — before anything sensitive is sent.
- [Browser Extension](https://docs.promptguard.co/shadow-ai/browser-extension.md): A roadmap enhancement for AI web apps — in-page coaching before you hit send. You don’t need it today: the desktop agent already covers the browser.
- [Fleet Enrollment](https://docs.promptguard.co/shadow-ai/fleet-enrollment.md): Roll Shadow AI out to your whole team — each device gets its own scoped, revocable credential, and you see all AI activity in one dashboard.
- [Deployment Modes](https://docs.promptguard.co/shadow-ai/deployment-modes.md): Decide where your prompts get scanned — in our cloud, on your own servers, or fully offline. Same dashboard either way.
- [Privacy & Data Handling](https://docs.promptguard.co/shadow-ai/privacy-data-handling.md): What Shadow AI inspects, what it logs, and what leaves the device. Local-first inspection, PII-redacted event logs, and GDPR/CCPA subject rights.
- [Known Limitations](https://docs.promptguard.co/shadow-ai/known-limitations.md): An honest list of what Shadow AI does not cover — certificate-pinned apps, QUIC/HTTP-3, remote images, the host-allowlist model — and how to request a vendor.
- [Troubleshooting](https://docs.promptguard.co/shadow-ai/troubleshooting.md): Fix common Shadow AI problems — traffic not intercepted, browser QUIC/HTTP-3 bypass, cert-pinned apps, engine unreachable — and how to collect support logs.
- [Uninstall & Offboarding](https://docs.promptguard.co/shadow-ai/uninstall.md): Cleanly remove the Shadow AI agent — drop the proxy, untrust the certificate, delete local data — and what happens when a device is revoked from the fleet.
- [Python SDK](https://docs.promptguard.co/guides/python-sdk.md): Secure every LLM call in your Python application with one line of code
- [Node.js SDK](https://docs.promptguard.co/guides/node-sdk.md): Drop-in security for AI applications -- auto-instrument OpenAI, Anthropic, Google, Cohere, and AWS Bedrock SDKs with a single line of code
- [Streaming](https://docs.promptguard.co/guides/streaming.md): Stream AI responses in real-time with PromptGuard security
- [Supported LLM Providers](https://docs.promptguard.co/guides/llm-providers.md): Complete list of all LLM providers and models supported by PromptGuard
- [Migration from OpenAI](https://docs.promptguard.co/guides/migration-from-openai.md): Step-by-step guide to migrate your existing OpenAI integration to PromptGuard
- [MCP Server](https://docs.promptguard.co/tools/mcp.md): Connect PromptGuard to any AI editor, coding agent, or app builder via the Model Context Protocol
- [CLI Tool](https://docs.promptguard.co/tools/cli.md): Scan codebases for unprotected LLM calls from your terminal
- [VS Code Extension](https://docs.promptguard.co/tools/vscode.md): PromptGuard for VS Code — see unprotected LLM calls inline as you type, with security warnings and one-click fixes in your editor.
- [Cursor Plugin](https://docs.promptguard.co/tools/cursor.md): AI-native LLM security for Cursor -- rules, commands, and MCP tools
- [GitHub Code Scanner](https://docs.promptguard.co/tools/github-scanner.md): Connect a GitHub repository and PromptGuard finds every unprotected LLM API call in it, then opens a pull request that adds protection.
- [Chatbot Protection](https://docs.promptguard.co/cookbooks/chatbot-protection.md): Secure your chatbot applications against prompt injection and abuse
- [Content Moderation](https://docs.promptguard.co/cookbooks/content-moderation.md): Implement comprehensive content filtering and moderation with PromptGuard
- [Data Privacy Protection](https://docs.promptguard.co/cookbooks/data-privacy.md): Implement comprehensive data privacy and PII protection with PromptGuard
- [LangChain Agents](https://docs.promptguard.co/cookbooks/langchain-agents.md): Secure LangChain agents and chains with PromptGuard
- [RAG with PII Redaction](https://docs.promptguard.co/cookbooks/rag-pii-redaction.md): Build a secure RAG pipeline that automatically redacts sensitive data
- [Streaming with Protection](https://docs.promptguard.co/cookbooks/streaming-protection.md): Secure streaming LLM responses in real time — scan server-sent event chunks for PII and jailbreaks without breaking the stream.
- [Managing Projects](https://docs.promptguard.co/platform/projects.md): Organize your applications with PromptGuard projects
- [Monitoring Dashboard](https://docs.promptguard.co/platform/dashboard.md): Access security metrics, usage analytics, and configure policies via the PromptGuard dashboard
- [Analytics Cookbook](https://docs.promptguard.co/platform/analytics-cookbook.md): Copy-pasteable SQL for your PromptGuard telemetry — p95 latency, token volume by model, high-risk users, blocked-over-time, threats by detector, and daily roll-ups.
- [Usage Tracking](https://docs.promptguard.co/platform/usage-tracking.md): Monitor API usage, costs, and consumption patterns with detailed analytics
- [Webhooks](https://docs.promptguard.co/platform/webhooks.md): Receive real-time notifications for security events
- [Audit Logs](https://docs.promptguard.co/platform/audit-logs.md): View security events, compliance audit trails, and interactions in your dashboard
- [Organizations & Teams](https://docs.promptguard.co/platform/organizations.md): Collaborate with your team using shared projects and role-based access
- [Single Sign-On (SSO)](https://docs.promptguard.co/platform/sso.md): Let your team sign in to PromptGuard with your identity provider via SAML or OIDC
- [Directory Sync (SCIM)](https://docs.promptguard.co/platform/scim.md): Automatically provision and deprovision PromptGuard members from your identity provider
- [Best Practices](https://docs.promptguard.co/production/best-practices.md): Production best practices for PromptGuard implementation
- [Reliability & Status](https://docs.promptguard.co/production/reliability.md): System health, uptime monitoring, and health check endpoints
- [Rate Limits & Quotas](https://docs.promptguard.co/production/rate-limits.md): Understanding PromptGuard rate limiting and monthly quotas
- [Latency Budgets](https://docs.promptguard.co/production/latency-budgets.md): Per-detector p99 latency budgets, the three-tier detection path, and how to keep the guard call fast in production.
- [Error Handling](https://docs.promptguard.co/production/error-handling.md): Comprehensive error handling strategies for PromptGuard integration
- [Enterprise Setup](https://docs.promptguard.co/production/enterprise-setup.md): Complete guide for deploying PromptGuard in enterprise environments with organizations, SSO, RBAC, compliance, and security controls
- [Self-Host vs Cloud](https://docs.promptguard.co/production/self-host-vs-cloud.md): An honest capability matrix of PromptGuard managed Cloud versus self-hosted and air-gapped — what is identical, what differs, why some parts are cloud-only.
- [Troubleshooting](https://docs.promptguard.co/production/troubleshooting.md): Common issues and solutions for PromptGuard integration
- [Glossary](https://docs.promptguard.co/glossary.md): Plain-language definitions of the security and product terms used throughout PromptGuard
- [Compliance & Security](https://docs.promptguard.co/security/compliance.md): Security certifications, data handling, and compliance information
- [Extended Threat Coverage](https://docs.promptguard.co/security/ai-agent-traps.md): Full list of threat types PromptGuard detects beyond standard prompt injection, including agentic, multimodal, and systemic attack vectors.
- [Plans & Limits](https://docs.promptguard.co/pricing.md): PromptGuard plan limits, quotas, rate limiting, and usage calculation
- [Changelog](https://docs.promptguard.co/changelog.md): Release notes for PromptGuard — new detectors, SDK updates, API changes, and bug fixes, newest first.
- [CI/CD Security Gate](https://docs.promptguard.co/tools/ci-cd-security-gate.md): Gate every pull request on AI security tests — run PromptGuard red-team checks in CI and block merges that regress security. Coming soon.
- [API Reference](https://docs.promptguard.co/api-reference/introduction.md): PromptGuard REST API reference — authentication, the Guard API, proxy endpoints, agent security, rate limits, and error handling.
- [Guard API](https://docs.promptguard.co/api-reference/guard.md): Scan content for threats without proxying to an LLM provider
- [Security Scan & Redact](https://docs.promptguard.co/api-reference/security-endpoints.md): Standalone endpoints for threat scanning and PII redaction
- [Agent Security API](https://docs.promptguard.co/api-reference/agent-security.md): Agent Security API — validate tool calls, register agents, ingest execution traces, and monitor agent behaviour for goal drift.
- [API Keys Management](https://docs.promptguard.co/api-reference/api-keys.md): Create, list, and manage API keys programmatically
- [Proxy Chat Completions](https://docs.promptguard.co/api-reference/proxy-chat-completions.md): Proxy OpenAI-style chat completions through policy engine
- [Proxy Messages](https://docs.promptguard.co/api-reference/proxy-messages.md): Proxy Anthropic-style messages endpoint through policy engine.
- [Proxy Responses](https://docs.promptguard.co/api-reference/proxy-responses.md): Proxy OpenAI's Responses API through the policy engine.
- [Proxy Count Tokens](https://docs.promptguard.co/api-reference/proxy-count-tokens.md): Proxy Anthropic's token-counting endpoint.
- [Proxy Models](https://docs.promptguard.co/api-reference/proxy-models.md): Proxy models endpoint.
- [Root](https://docs.promptguard.co/api-reference/root.md): API root endpoint.
- [Health](https://docs.promptguard.co/api-reference/health.md): Health check endpoint.
- [Version](https://docs.promptguard.co/api-reference/version.md): Public version probe.
- [List Api Keys](https://docs.promptguard.co/api-reference/api-keys/list-api-keys.md): List all API keys for the current user
- [Create Api Key](https://docs.promptguard.co/api-reference/api-keys/create-api-key.md): Create a new API key
- [Delete Api Key](https://docs.promptguard.co/api-reference/api-keys/delete-api-key.md): Delete an API key
- [Toggle Api Key](https://docs.promptguard.co/api-reference/api-keys/toggle-api-key.md): Toggle API key active status
- [Reveal Api Key](https://docs.promptguard.co/api-reference/api-keys/reveal-api-key.md): Reveal the full API key for copying.
- [Get Usage Stats](https://docs.promptguard.co/api-reference/usage/get-usage-stats.md): Get current user's usage statistics
- [List Projects](https://docs.promptguard.co/api-reference/projects/list-projects.md): List all projects for the current user
- [Create Project](https://docs.promptguard.co/api-reference/projects/create-project.md): Create a new project
- [Get Project](https://docs.promptguard.co/api-reference/projects/get-project.md): Get a specific project
- [Delete Project](https://docs.promptguard.co/api-reference/projects/delete-project.md): Delete a project
- [Validate Tool Call](https://docs.promptguard.co/api-reference/agent/validate-tool-call.md): Validate a tool call before execution.
- [Ingest Agent Trace](https://docs.promptguard.co/api-reference/agent/ingest-agent-trace.md): Ingest a full agent execution trace and run the trace-level detectors.
- [Scan Agent Memory](https://docs.promptguard.co/api-reference/agent/scan-agent-memory.md): Scan a chunk of agent memory for poisoning (OWASP Agentic ASI06).
- [Get Agent Stats](https://docs.promptguard.co/api-reference/agent/get-agent-stats.md): Get statistics for an agent.
- [End Agent Session](https://docs.promptguard.co/api-reference/agent/end-agent-session.md): Deprecated no-op. Agent session state is not retained server-side across requests, so there is nothing to end. Returns 200 for compatibility; scheduled for removal — see the Sunset header.
- [Register Agent](https://docs.promptguard.co/api-reference/agent/register-agent.md): Register a new agent and return a one-time-visible credential.
- [Rotate Agent Credential](https://docs.promptguard.co/api-reference/agent/rotate-agent-credential.md): Revoke the current credential and issue a new one.
- [Agent Security Health](https://docs.promptguard.co/api-reference/agent/agent-security-health.md): Health check for agent security service.
- [Managed Policy](https://docs.promptguard.co/api-reference/agent/managed-policy.md): The org-managed update policy for this enrolled device (resolved from the API key's org + its ``shadow_ai_fleet`` entitlement). The desktop agent polls this and lets it override the local user preference (managed wins).
- [Scan Content](https://docs.promptguard.co/api-reference/security/scan-content.md): Scan content for security threats without proxying to an LLM.
- [Redact Content](https://docs.promptguard.co/api-reference/security/redact-content.md): Redact PII from content without proxying to an LLM.
- [Guard Content](https://docs.promptguard.co/api-reference/guard/guard-content.md): Scan messages for security threats without proxying to an LLM.
- [Enroll Device](https://docs.promptguard.co/api-reference/enroll/enroll-device.md)
- [List Exceptions](https://docs.promptguard.co/api-reference/shadow-exceptions/list-exceptions.md)
- [Create Exception](https://docs.promptguard.co/api-reference/shadow-exceptions/create-exception.md)
- [List Active Grants](https://docs.promptguard.co/api-reference/shadow-exceptions/list-active-grants.md): The destinations this agent may currently send to via an approved grant. The interceptor caches these and lets matching sends through.
- [Get Exception](https://docs.promptguard.co/api-reference/shadow-exceptions/get-exception.md)
- [Cancel Exception](https://docs.promptguard.co/api-reference/shadow-exceptions/cancel-exception.md)
- [List Policies](https://docs.promptguard.co/api-reference/policies/list-policies.md): Active policies enforced on this device: the project's own + the account's global (project-less) policies, highest-priority first.
- [Preview Overlay](https://docs.promptguard.co/api-reference/overlays/preview-overlay.md): Evaluate a candidate overlay in shadow against a traffic sample.
- [Apply Overlay Endpoint](https://docs.promptguard.co/api-reference/overlays/apply-overlay-endpoint.md): Promote the overlay to active as a new version for the key's scope.
- [Get Active Overlay](https://docs.promptguard.co/api-reference/overlays/get-active-overlay.md): Return the current active overlay for the key's scope, if any.
- [List Tool Requests](https://docs.promptguard.co/api-reference/tool-requests/list-tool-requests.md)
- [Create Tool Request](https://docs.promptguard.co/api-reference/tool-requests/create-tool-request.md)
- [Cancel Tool Request](https://docs.promptguard.co/api-reference/tool-requests/cancel-tool-request.md)
- [Handle Webhook](https://docs.promptguard.co/api-reference/github-webhooks/handle-webhook.md): Receive and process GitHub App webhook events.

## OpenAPI Specs

- [openapi-developer](https://docs.promptguard.co/api-reference/openapi-developer.json)
- [openapi](https://docs.promptguard.co/api-reference/openapi.json)

## Optional

- [Dashboard](https://app.promptguard.co)
- [Blog](https://promptguard.co/blog)