# PromptGuard

> PromptGuard is a freemium MCP server for security. Scan prompts for injection and PII, and audit LLM SDK usage from an MCP client. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API, self-hosted, CLI
- **Website:** https://promptguard.co
- **Repository:** https://github.com/promptguardhq/promptguard
- **MCP Registry name:** co.promptguard/security
- **Categories:** security, AI models and LLM tooling
- **Listing updated:** 2026-11-14

## About

PromptGuard sits in front of the model as a firewall, and this server lets an MCP client reach the same detection engine to scan prompts for injection attacks, redact personal data and audit how an LLM SDK is used. It needs a PromptGuard API key.

The vendor publishes its own measurements: on 2,369 samples across eight datasets it reports 99.1% precision and 80.3% recall, and it says openly that over-defense on a hard benign set is 45.4%. It lists 15 detectors and 46 PII entity types, with a fast path under 10 ms for deterministic checks.

The engine can run in the vendor's cloud, hybrid in your own infrastructure with Helm or docker-compose, or air-gapped. Zero-retention and metadata-only logging modes are described, and the site states it holds no SOC 2, ISO 27001 or FedRAMP certification.

## Features

- **Injection scanning:** Checks prompts for injection and jailbreak attempts before they reach a model.
- **PII redaction:** Detects and redacts 46 personal data entity types.
- **SDK usage audit:** Audits how LLM SDKs are used in a codebase.
- **Three deployment modes:** Managed cloud, hybrid self-host or fully air-gapped.
- **Published benchmarks:** Reports precision, recall and over-defense with datasets and confidence intervals.

## Good for

- Screen user input in an agent before it is acted on
- Redact personal data from prompts
- Review a project's LLM SDK calls for risky usage
- Add a guard layer inside a private network

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http security https://api.promptguard.co/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.promptguard.co/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "security": {
      "url": "https://api.promptguard.co/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "security": {
      "type": "http",
      "url": "https://api.promptguard.co/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "security": {
      "serverUrl": "https://api.promptguard.co/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "security": {
      "type": "http",
      "url": "https://api.promptguard.co/mcp"
    }
  }
}
```

## Tools

- `promptguard_scan_text`: Scan text for security threats including prompt injection, jailbreaks, PII leakage, and toxic content. Returns a decision (allow/block), confidence score, threat type, and reason. Each call counts as one Request on your PromptGuard plan. When to use: When reviewing user-facing prompts, testing LLM inputs for safety, or verifying that content is free of injection attacks before sending to an LLM.
- `promptguard_scan_code`: Analyze code for LLM SDK usage patterns (OpenAI, Anthropic, Cohere, Google Gemini, Bedrock). Detects import statements and client instantiations via pattern matching. When to use: After generating new code that uses LLM SDKs, when reviewing pull requests, or when auditing a codebase for security coverage gaps.
- `promptguard_redact`: Redact PII from text. Returns the text with each entity replaced by a labeled placeholder, e.g. [EMAIL_REDACTED], [PHONE_REDACTED], [SSN_REDACTED], the PII types found, and `skipped`: the matches deliberately left in the text. Never returns the original text. Each call counts as one Request on your PromptGuard plan. Without `pii_types` it redacts what the Project's policy redacts; a default Projec
- `promptguard_status`: Show whether PromptGuard is set up and signed in, with this server's version and the account or configuration details it knows (the hosted server reports your plan; the local CLI server reports its key type, proxy URL and providers). When to use: To verify PromptGuard is properly configured, or to confirm which version is running.
- `promptguard_list_events`: List your organization's most recent PromptGuard security events: what was blocked, redacted, flagged or allowed, with the threat types, confidence, direction, surface, provider, project id and each finding's detector and rule. Newest first, at most 50. Filter by decision, threat type, project (by id, or by the name the User gave) or a start time. Only PromptGuard's own values are returned: never 
- `promptguard_get_usage`: Show this billing month's PromptGuard usage against your plan: Requests used and remaining, the monthly limit, usage credit where the plan has one, and the days until it resets. Read-only. When to use: To check whether the account is close to its plan limit.
- `promptguard_list_policies`: List your organization's PromptGuard security policies across every project: name, type, whether it is active and enforced, its priority and threat types, and how often it has triggered. Read-only; it cannot change a policy. Names and descriptions in the result are free text your organization recorded in PromptGuard. When to use: To see which protections are configured before explaining a block, o
- `promptguard_list_audit_log`: List recent entries in your organization's PromptGuard audit trail (who changed what, sign-ins, key and policy changes), newest first, at most 50. Filter by category, action, severity, outcome or a start time. Needs a plan with audit logs. Read-only. Names and descriptions in the result are free text your organization recorded in PromptGuard. When to use: To answer who changed a setting or key, an
- `promptguard_list_projects`: List the PromptGuard projects you are shown in your organization and how each is configured: strictness, fail mode, enforcement mode, zero retention, agent policy enforcement, and whether alerts go to a webhook or alert channels (the kind, never the URL). No keys or secrets. Read-only. Names and descriptions in the result are free text your organization recorded in PromptGuard. When to use: To che
- `promptguard_get_fleet_status`: Show your organization's Shadow AI fleet: whether Shadow AI is required, its mode and fail-closed setting, how many Devices are active or revoked, and for each Device its platform, agent version, coverage, proxy state and when it was last seen. Needs an organization admin. No enrollment tokens or keys. Read-only. Names and descriptions in the result are free text your organization recorded in Prom
- `promptguard_list_red_team_tests`: List the attacks PromptGuard's security testing (red team) runs against a project's guardrails: each test's name, category, what it tries and the result expected, and whether your organization's plan includes running them. It lists the tests; it runs nothing. Read-only. When to use: To explain what security testing covers before running it.
- `promptguard_list_webhook_deliveries`: List recent alert deliveries to your projects' webhooks and alert channels, newest first, at most 50: whether each was delivered, is pending or failed, its attempts and the receiver's HTTP status. Never the payload, the URL or the receiver's response. Read-only. When to use: To check whether alerts are reaching their destination.
- `promptguard_propose_suppression`: Suppress one rule of one detector on one project until it expires: the rule stops deciding the project's requests, and its findings are still recorded, marked suppressed. Name the project by its id and the rule as a finding names it (rule.detector and rule.id, e.g. from an event). Confirmation: in the PromptGuard dashboard this returns a Proposal the User confirms or dismisses in the panel, within

## FAQ

### Does it need an account?

Yes, an API key starting with pg_live_ is required.

### Can it run without sending data out?

Yes. A hybrid self-host and an air-gapped option are described.

### Is it certified?

The site says it holds no SOC 2 Type II, ISO 27001 or FedRAMP authorization.

### Is PromptGuard free?

PromptGuard has a free plan, with paid plans for more usage or features.

### How do I add PromptGuard to Claude Code?

Run this command in your terminal: claude mcp add --transport http security https://api.promptguard.co/mcp

### What are the best alternatives to PromptGuard?

The closest alternatives to PromptGuard are REA, Claude Code Ultimate Guide, Agent Roast, Midplane and Leash.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Agent Roast](https://ai-powered.com/mcp/agent-roast): Agent Roast is an open-source MCP server for developer tools.
- [Midplane](https://ai-powered.com/mcp/midplane): Midplane is an open-source MCP server for databases.
- [Leash](https://ai-powered.com/mcp/leash): Leash is a free MCP server for security.
- [Tinplate Edge AI Safety, Security & EU Compliance](https://ai-powered.com/mcp/tinplate-edge-ai-safety-security-and-eu-compliance): Tinplate Edge AI Safety, Security & EU Compliance is a freemium MCP server for security.
- [SecureCloud](https://ai-powered.com/mcp/securecloud): SecureCloud is a free MCP server for security.
- [Eikino](https://ai-powered.com/mcp/eikino): Eikino is a freemium MCP server for security.

---
Source: https://ai-powered.com/mcp/promptguard
