Fabraix Red Team Playground
Free · Open source
AI security researchers
AxioRank
FreemiumAPI
Production agents running with live cloud credentials are one bad tool call away from a deleted table or an exfiltrated key — and most frameworks give you no checkpoint between the model and your infrastructure.
$49/mo · Free tier: 10K events/month, 2 agents, 2 team seats, 10 requests/second, 7 days audit retention · Checked Jul 1, 2026
Similar tools
Fabraix Red Team PlaygroundFreeFabraix Red Team Playground is an open-source, MIT-licensed testing environment structured…
EclusiaFreemiumEclusia runs locally on your machine and replaces names, IBANs, phone numbers, API keys, and…
Patronus ScannerFreemiumPatronus AI positions itself as evaluation and monitoring infrastructure for production LLM…
ArgotFreeArgot runs statistical checks against your repository's own commit history to flag code that…
Summary
AxioRank sits between your agents and every surface they can reach — MCP servers, APIs, databases — and runs a verify-score-decide-record pipeline on every tool call before anything executes. Short-lived tokens default to a 15-minute lifetime, so a leaked credential expires before it causes damage. Thirty-one content detectors score each payload for credential leakage, destructive SQL, and SSRF attempts, and the policy engine resolves deny-overrides in under 100 ms on the synchronous path. The vendor states secrets are fingerprinted but never stored, and audit rows are redacted and append-only. SSO and extended audit retention are paid-only features, so teams with strict compliance requirements hit that wall fast.
Bottom line: Pick AxioRank when you need a gateway between autonomous agents and production infrastructure without rewriting your agents — but plan for a paid tier upgrade the moment your security team asks for SSO or long-retention audit exports.
Pricing Plans
SubscriptionLast verified 4 months ago- Price
- $49/mo
- Free Tier
- 10K events/month, 2 agents, 2 team seats, 10 requests/second, 7 days audit retention
Free
Free
For trying it out and small projects.
- 10K events/month
- 2 agents
- 2 team seats
- 10 requests/second
- 7 days audit retention
- Monitor-mode response
- Full protocol coverage
Pro
$49per month
For teams running agents in production.
- 250K events/month
- 10 agents
- 3 team seats
- 50 requests/second
- 30 days audit retention
- Anomaly detection
- Armed response
- Custom detectors
Team
$299per month
For security teams that need depth.
- 2M events/month
- 50 agents
- 10 team seats
- 200 requests/second
- 90 days audit retention
- Model threat intel
- Secrets broker
- Audit export
- SAML SSO (early access)
Enterprise
Custom
For organizations with scale and compliance needs.
- Unlimited events/month
- Unlimited agents
- Unlimited team seats
- Unlimited requests/second
- 1 year audit retention
- Uptime SLA
- One-year retention
- Unlimited volume
View full pricing on axiorank.com →
Pricing may have changed since last verified. Check the official site for current plans.
Find your fit
Before you choose AxioRank
More options in Guardrails & Safety. Compare pricing, platforms, and features side by side.
Eclusia
Freemium
Legal professionals handling confidential cases
Pros
- Short-lived tokens with a 15-minute default lifetime, so a credential leak from a tool call expires before an attacker can replay it — without requiring your agents to be rewritten.
- 31 payload detectors scanning for credential exposure, destructive SQL, and SSRF on every tool call, which means a model generating a DELETE without a WHERE clause or pointing an agent at the cloud metadata endpoint gets caught before execution.
- Deny-overrides policy engine returning a decision in under 100 ms on the synchronous path, so enforcement does not require async sidecars or post-execution callbacks — the agent never hears back if the call is denied.
- Default-deny tool and egress allowlist, so any tool or external endpoint your agent calls that you have not explicitly approved is blocked — closing the gap that permissive-by-default frameworks leave open.
- Redacted append-only audit rows on every tool call, so your compliance team has a signed trail of what every agent did and when, without secrets appearing in the log.
Cons
- The default-deny allowlist requires enumerating every tool call and egress destination before the gateway can run in production. Teams with agents that make ad-hoc or dynamically generated API calls spend a significant configuration phase building that list — and any undocumented call path the agent needs gets blocked on first contact.
- SSO and extended audit retention are paid-only features. Security teams at organizations where SSO is a non-negotiable access control requirement hit this wall immediately and must upgrade before they can evaluate the tool against their compliance checklist.
- There is no self-hosted option. Teams in environments where all security tooling must run inside their own perimeter — regulated industries, air-gapped infrastructure — cannot deploy AxioRank and move to a competitor or build an in-house gateway instead.
- No named competitors in the market segment means teams vetting this tool have no established benchmark for comparison, which extends the evaluation cycle for procurement teams required to document alternatives before approving spend.
About
- API Available
- Yes
- Self-Hosted
- No
- Last Updated
- June 30, 2026
Best For
Who it's for
- Teams running AI agents in production
- Security teams needing anomaly detection and response
- Organizations requiring audit retention and SSO
What it does well
- Securing agent access to MCP servers, APIs, and databases
- Enforcing policies and inspecting content on agent tool calls
- Generating signed audit trails for compliance
- Verifying agents visiting external sites
Integrations
Frequently Asked Questions
- Is AxioRank free?
- AxioRank offers a free option alongside paid upgrades. Pricing: $49/mo. Check the current allowances and terms before choosing a plan.
- Does AxioRank have an API?
- Yes. AxioRank exposes a developer API. See the official documentation at https://axiorank.com for details.
Curated lists that include this category
Give a model tools and it can delete a database row, call your cloud provider, or carry a live API key straight to an external endpoint. AxioRank is a security gateway that sits on the path between agents and everything they touch. Every tool call passes through a four-stage pipeline: identity verification using short-lived signed tokens that resolve locally without a database round-trip, payload inspection across 31 detectors that score risk from 0 to 100, policy evaluation under deny-overrides logic that returns allow, deny, or hold, and a redacted append-only audit record. The vendor states the entire decision completes in under 100 ms on the synchronous path, so agents are not rewritten and latency stays low.
The identity model is the clearest differentiator. Rather than trusting agents with static long-lived secrets, AxioRank issues short-lived signed tokens with a default 15-minute lifetime. A leaked token is worthless within minutes. The gateway also supports federated identity exchange, so agents already authenticated through another system do not need a separate credential store.
AxioRank integrates with the frameworks teams already run — the vendor lists MCP, LangChain, LangGraph, OpenAI Agents SDK, Anthropic SDK, CrewAI, AutoGen, LiteLLM, and others. The default-deny tool and egress allowlist means anything not explicitly approved is blocked, not just flagged. This is the right posture for production, but it also means initial setup requires enumerating every tool call and external endpoint your agents use — teams with poorly documented agent behavior spend meaningful time building that allowlist before they can cut traffic through the gateway. SSO and extended audit retention are paid-only features, which is the point at which compliance-focused organizations must upgrade or look elsewhere.
Contribute
- Work on AxioRank? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.