Freedom251 › AI Agent Security Software
GuardionAI
What GuardionAI is
GuardionAI provides runtime security and governance for customer-facing AI, autonomous agents, and coding agents. Its gateway observes and evaluates agent and MCP calls inline, then can allow, block, or redact actions before execution. Data-loss prevention identifies personal information and secrets, masking or vault-tokenizing them before requests leave the organization and restoring responses for users. Runtime guardrails detect prompt injection, jailbreaks, harmful content, and unsafe tool calls, with options to block, redact, or flag actions. A coding-agent plugin can stop destructive commands, secret exfiltration, and unauthorized tool calls before they run. The service can operate inline as a gateway or through its API without application rewrites. Integrations listed include OpenAI, Claude, Gemini, LangChain, LlamaIndex, LangGraph, Pinecone, AWS, Google Cloud, and Microsoft Azure. GuardionAI says its gateway is encrypted, providers do not see raw data, and customer data is not used to train its models. A free trial is offered, but no duration is stated.
Who it suits
GuardionAI is aimed at organizations securing AI applications, autonomous agents, and coding-agent workflows. It may suit teams needing inline controls, DLP, guardrails, and exportable audit logs.
What it does well
- Blocks or redacts unsafe agent actions
- Masks PII and secrets before requests leave
- Coding-agent protection against destructive commands
- API or inline gateway deployment
- Free trial available
What to know first
- No free plan is listed
- Pay-as-you-go retains logs for seven days
- Enterprise pricing is custom
Limits
Pay-as-you-go includes one seat, seven-day log retention, one million requests per month, and 1,000 requests per minute; guardrail scans and observability requests are pay-per-use. Enterprise has unlimited requests and seats, up to 365-day log retention, and custom limits under a custom contract.
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Enterprise | Custom contract · Unlimited seats and requests · up to 365-day log retention · custom rate limit · 24/7 priority support · 99.99% SLA · on-premises deployment | On request |
| Pay-as-you-go | Pay-per-use · 1 seat · 7-day log retention · 1M requests/month · 1,000 requests/minute · community support · 99.5% SLA | On request |
As guardion.ai lists them, read 21 Oct 2026. The maker's own page has the final word.
Getting started
- Open https://guardion.ai/
- Start the free trial
- Choose API, gateway, or a supported deployment environment
- Connect relevant AI providers, frameworks, or cloud services
- Select Pay-as-you-go or Enterprise
- Configure policies for guardrails, data protection, and tool calls
Verdict
GuardionAI focuses on inline controls for AI calls, sensitive data, and coding-agent tools. Its available plans differ in retention, request limits, support, and pricing terms.
Questions
Does GuardionAI have a free plan?
No. GuardionAI is paid, but the site offers a free trial that can be canceled anytime.
Which platforms can run GuardionAI?
It supports API, Linux, macOS, self-hosted, web, and Windows environments.
What does the gateway protect?
It evaluates agent and MCP calls inline and can allow, block, or redact actions before execution.
Which integrations are listed?
Listed support includes OpenAI, Claude, Gemini, LangChain, LlamaIndex, LangGraph, Pinecone, AWS, Google Cloud, Microsoft Azure, and other AI services.
What are the Pay-as-you-go limits?
The plan includes one seat, seven-day log retention, one million requests per month, and 1,000 requests per minute. Guardrail scans and observability requests are charged per use.
Who makes GuardionAI?
The website identifies Metatext AI LLC as the company behind GuardionAI.
Facts from GuardionAI's own pages
- API products
- The API documentation exposes a Guard evaluation API at api.guardion.ai and an OpenAI-compatible Gateway at gateway.guardion.ai. · guardion.ai, 21 Oct 2026
- API protocol
- Guard endpoints use versioned JSON APIs, with Guard under /v1 and findings also using /v2. · guardion.ai, 21 Oct 2026
- Audit
- Audit logs are described as tamper-evident and exportable to a SIEM. · guardion.ai, 29 Sept 2026
- Coding-agent controls
- Claude Code plugins monitor coding agents and can block destructive commands, credential leaks and risky changes. · guardion.ai, 21 Oct 2026
- Coding agents
- The coding-agent plugin monitors tool calls and can block destructive commands, secret exfiltration, and unauthorized tool calls before they run. · guardion.ai, 29 Sept 2026
- Company
- The website identifies Metatext AI LLC as the company behind GuardionAI. · guardion.ai, 29 Sept 2026
- Compliance
- GuardionAI describes itself as GDPR-, HIPAA-, and LGPD-ready and says SOC 2 Type II is in progress. · guardion.ai, 29 Sept 2026
- Compliance status
- GuardionAI describes itself as GDPR, HIPAA and LGPD ready, while SOC 2 Type II is still in progress. · guardion.ai, 21 Oct 2026
- Crunchbase profile
- Crunchbase describes Guardion.AI as a private, active company providing runtime security and observability for AI agents and lists its legal name as Guardion.AI. · crunchbase.com, 21 Oct 2026
- Data handling
- The company says its gateway is encrypted, model providers do not see raw data, customer data is not used to train its models and PII is stripped before leaving the organization. · guardion.ai, 21 Oct 2026
- Data-loss prevention
- The product detects and redacts personal information and secrets before they leave an organization. · guardion.ai, 21 Oct 2026
- Data protection
- The DLP feature identifies PII and secrets, masks or vault-tokenizes them before requests leave the organization, and can restore responses for users. · guardion.ai, 29 Sept 2026
- Deployment
- GuardionAI says it can run inline as a gateway or be called through its API, with no application rewrites. · guardion.ai, 29 Sept 2026
- Deployment speed
- GuardionAI says deployment takes under 30 minutes and requires zero instrumentation. · guardion.ai, 21 Oct 2026
- Detection coverage
- Listed detections include prompt attacks, PII, secrets, content moderation, topic drift, MCP scope violations and anomalies. · guardion.ai, 21 Oct 2026
- Documentation depth
- Public documentation includes API reference pages, cURL examples, detector descriptions, policy behavior, telemetry schemas and DLP integration guidance. · guardion.ai, 21 Oct 2026
- Enterprise deployment
- Enterprise pricing includes on-premises deployment, custom data regions, SSO, SCIM and granular RBAC. · guardion.ai, 21 Oct 2026
- Funding report
- OnStage VC's profile reports GuardionAI had raised between £150,000 and £250,000 and names David Holanda and Patrick Sigrits among investors. · onstage.vc, 21 Oct 2026
- Gateway
- Its gateway observes and evaluates agent and MCP calls inline, then can allow, block, or redact actions before execution. · guardion.ai, 29 Sept 2026
- Gateway compatibility
- The Gateway works by changing an existing SDK base URL and adding Guardion headers while forwarding the customer's provider key. · guardion.ai, 21 Oct 2026
- Guard models
- The documented model lineup includes prompt-defense detection for injection, jailbreak, spam and bot abuse, with scores, labels and thresholds. · guardion.ai, 21 Oct 2026
- Guardrails
- Runtime guardrails detect prompt injection, jailbreaks, harmful content, and unsafe tool calls, and can block, redact, or flag actions. · guardion.ai, 29 Sept 2026
- Incident response
- A centralized console provides runtime monitoring, findings, tracing and investigation workflows for agent incidents. · guardion.ai, 21 Oct 2026
- Industry recognition
- GuardionAI's LinkedIn page says it was featured in OWASP's Q2 2026 AI Security Solutions Landscape for Agentic AI. · linkedin.com, 21 Oct 2026
Where GuardionAI stands on Freedom251
- AI Agent Security Software#8 of 31
- LLM Security Tools#12 of 31