PromptGuard changelog

• + promptguard_list_events
• + promptguard_get_usage
• + promptguard_list_policies
• + promptguard_list_audit_log
• + promptguard_list_projects
• + promptguard_get_fleet_status
• + promptguard_list_red_team_tests
• + promptguard_list_webhook_deliveries
• + promptguard_propose_suppression

https://promptguard.co/llms.txt

− - Standalone Python MCP server: `pip install promptguard-mcp-server` or `uvx promptguard-mcp-server` or `npx @promptguard/mcp-server` or `docker run abhijoysarkar/promptguard-mcp-server`− - Also available as native MCP server built into the Rust CLI: `promptguard mcp -t stdio`− - Supports stdio and Streamable HTTP transports− - Exposes 6 tools: auth, logout, scan_text, scan_project, redact, status− - Cursor plugin available at https://github.com/promptguardhq/promptguard-cursor− # ChatGPT App & Codex Plugin− - HTTPS MCP server at https://api.promptguard.co/mcp for ChatGPT Apps SDK− - 7 tools: promptguard_scan_text, promptguard_scan_code, promptguard_redact, promptguard_status, search, fetch− - OAuth 2.1 authentication via Auth0 (PKCE + DCR)− - ChatGPT App docs: https://docs.promptguard.co/tools/chatgpt− Developers and engineering teams building LLM-powered applications who need enterprise-grade security without rewriting their code.− - SOC 2 Type II: not certified. Not started; no audit is scheduled.− - Cursor Plugin: https://github.com/promptguardhq/promptguard-cursor+ - MCP docs: https://docs.promptguard.co/tools/mcp+ - VS Code extension (also runs in Cursor as a .vsix): Marketplace ID promptguard.promptguard-vscode, marks unprotected LLM SDK calls and scans a selection or file; needs the PromptGuard CLI and an API key set with "PromptGuard: Set API Key". Results go to the PromptGuard output channel - https://docs.promptguard.co/tools/vscode+ - Agent plugin for Claude Code, Codex and Cursor (skills, a security reviewer, Cursor rules and the hosted MCP server) - https://github.com/promptguardhq/promptguard-plugin+ ## Hosted MCP server (also the ChatGPT App and Codex plugin)+ - Streamable HTTP at https://api.promptguard.co/mcp; sign in with your PromptGuard account (OAuth 2.1 via Auth0, PKCE + DCR), no API key to copy+ - Hosted server tools (13): promptguard_scan_text, promptguard_redact, promptguard_status, promptguard_scan_code, promptguard_list_events, promptguard_get_usage, promptguard_list_policies, promptguard_list_audit_log, promptguard_list_projects, promptguard_get_fleet_status, promptguard_list_red_team_tests, promptguard_list_webhook_deliveries, promptguard_propose_suppression+ ## Local MCP server (built into the CLI)+ - stdio only, authenticated with your PromptGuard API key: `promptguard mcp -t stdio`+ - Same server, launched another way: `npx @promptguard/mcp-server`, or `docker run -i --rm -e PROMPTGUARD_API_KEY ghcr.io/promptguardhq/promptguard-mcp-server` (linux/amd64). The older abhijoysarkar/promptguard-mcp-server image on Docker Hub was removed on 2026-11-14.+ - Local server tools (6): promptguard_scan_text, promptguard_redact, promptguard_status, promptguard_auth, promptguard_logout, promptguard_scan_project+ Developers and engineering teams building LLM-powered applications who need their LLM traffic checked without rewriting their code.+ - SOC 2: not certified. Type 1 preparation started October 2026; no report held.+ - VS Code extension: https://docs.promptguard.co/tools/vscode+ - Agent plugin (Claude Code, Codex, Cursor): https://github.com/promptguardhq/promptguard-plugin+ - Agent setup (for AI agents to follow): https://promptguard.co/get-started.md

version

1.2.4 → 1.2.5

version

1.2.3 → 1.2.4

icon_url

— → https://promptguard.co/logo.svg

version

1.2.1 → 1.2.3

website_url

— → https://promptguard.co

version

1.2.0 → 1.2.1

remote_url

— → https://api.promptguard.co/mcp

transport

stdio → streamable-http

version

1.1.3 → 1.2.0

https://promptguard.co/llms.txt

− - Free tier: 20,000 requests/month− - Python SDK: promptguard-sdk on PyPI (MIT license, open source) - https://github.com/acebot712/promptguard-python− - Node.js SDK: promptguard-sdk on npm (MIT license, open source) - https://github.com/acebot712/promptguard-node− - CLI: promptguard-cli (Apache 2.0, Rust), installed with its install script or a prebuilt binary from GitHub Releases - https://github.com/acebot712/promptguard-cli− - Cursor plugin available at https://github.com/acebot712/promptguard-cursor− - Free: $0/month, 20,000 requests, prompt injection + PII detection (ML-enhanced); the full detector set starts on Pro− - Pro: $99/month, 100K requests, full detector set (exfiltration, toxicity, fraud, secrets, malware, URL filtering), custom policies + email alerts− - Scale: $199/month, 500,000 requests included, then metered overage at $0.40 per 1,000 up to a spend cap (default $500); past the cap requests return 429. Advanced analytics.− - SOC 2 Type II: not yet certified, on the roadmap. No audit has been completed.− - ISO 27001: not held; on the roadmap, after SOC 2. EU AI Act and ISO/IEC 42001: technical controls aligned, not certified.− - Access control: SAML/OIDC SSO, SCIM directory sync, four-role RBAC, and audit logging mapped to SOC 2 CC6.1/CC7.2/CC8.1.− - Uptime target 99.9%. This is a target, not a contractual SLA; no signed SLA is offered and there is no public status page.− - CLI: https://github.com/acebot712/promptguard-cli− - Cursor Plugin: https://github.com/acebot712/promptguard-cursor+ - Free tier: $5 of usage a month at list prices+ - Python SDK: promptguard-sdk on PyPI (MIT license, open source) - https://github.com/promptguardhq/promptguard-python+ - Node.js SDK: promptguard-sdk on npm (MIT license, open source) - https://github.com/promptguardhq/promptguard-node+ - CLI: promptguard-cli (Apache 2.0, Rust), installed with its install script or a prebuilt binary from GitHub Releases - https://github.com/promptguardhq/promptguard-cli+ - Cursor plugin available at https://github.com/promptguardhq/promptguard-cursor+ - Free: $0/month, $5 of usage a month at list prices (about 4,500 calls before tokens), the full detector set on the same engine as every plan+ - Developer: $20/month, billed monthly; the whole fee is $20 of usage a month at list prices, then list prices up to a spend cap (default $100, up to $10,000); past the cap requests return 429. Full detector set (exfiltration, toxicity, fraud, secrets, malware, URL filtering), custom policies, advanced analytics, email and webhook alerts+ - List prices: $1.10 per 1,000 calls + $0.55 per 1M billed tokens, on every plan: one detection engine, both LLM judges included, for everyone+ - SOC 2 Type II: not certified. Not started; no audit is scheduled.+ - ISO 27001: not certified. Not started; no audit is scheduled. EU AI Act and ISO/IEC 42001: technical controls aligned, not certified.+ - Access control: SAML/OIDC SSO, four-role RBAC, and audit logging mapped to SOC 2 CC6.1/CC7.2/CC8.1.+ - Uptime target 99.9%. This is a target, not a contractual SLA; no signed SLA is offered. Public status page: https://status.promptguard.co+ - CLI: https://github.com/promptguardhq/promptguard-cli+ - Cursor Plugin: https://github.com/promptguardhq/promptguard-cursor

repository_url

https://github.com/acebot712/promptguard → https://github.com/promptguardhq/promptguard

version

1.1.2 → 1.1.3

https://promptguard.co/llms.txt

− - PII detection & redaction (43 entity types with synthetic data replacement)− - CLI: promptguard-cli on crates.io and Homebrew (Apache 2.0, Rust) - https://github.com/acebot712/promptguard-cli− - Vulnerability disclosure: security@promptguard.co, 48-hour acknowledgement. See https://promptguard.co/.well-known/security.txt+ - PII detection & redaction (46 entity types with synthetic data replacement)+ - CLI: promptguard-cli (Apache 2.0, Rust), installed with its install script or a prebuilt binary from GitHub Releases - https://github.com/acebot712/promptguard-cli+ - Vulnerability disclosure: security@promptguard.co. Policy (timeline, scope, safe harbour, credit): https://promptguard.co/security#reporting-a-vulnerability

https://promptguard.co/llms.txt

− - Hosted service runs in Google Cloud us-central1. There is no hosted EU region; data residency is achieved by self-hosting.+ - Hosted service runs in Google Cloud us-east4 (Northern Virginia). There is no hosted EU region; data residency is achieved by self-hosting.

https://promptguard.co/llms.txt

− - ISO 27001: planned for 2026. EU AI Act and ISO/IEC 42001: technical controls aligned, not certified.+ - ISO 27001: not held; on the roadmap, after SOC 2. EU AI Act and ISO/IEC 42001: technical controls aligned, not certified.