Catch leaked secrets, prompt injection, and destructive calls in your AI agent's tool calls. Runs locally with no signup; enforce centrally with AxioRank.

AxioRank guardrail middleware for the Vercel AI SDK. Inspect and redact prompts, completions, and proposed tool calls on every generateText and streamText.

AxioRank CLI: export, diff, and import workspace governance as code.

AxioRank content-inspection engine: detectors + risk scoring for AI agent tool calls. Catches secrets, prompt injection, PII, destructive ops, egress, malware and crypto/financial identifiers. Pure, universal, zero runtime dependencies.

AxioRank nodes for n8n. Trigger on AI agent security events and act on them: approve or deny holds, quarantine agents, and update alerts and incidents.

Transparent MCP gateway: route an MCP server's tool calls through AxioRank for inspection, redaction, and policy enforcement. Zero-Trust governance by changing one URL.

A self-hostable witness for an AxioRank audit transparency log. Countersigns each sealed signed tree head with an Ed25519 key you control, so your records are an independent split-view defense. Zero runtime dependencies.

Independently monitor an AxioRank audit transparency log for equivocation, rollback, or gaps. Pins each verified signed tree head and alerts on any inconsistency. Zero runtime dependencies beyond the offline verifier.

AxioRank for Wix: verify and govern the AI agents that reach a Wix site. A thin client of the AxioRank inbound verify endpoint, with an Astro request middleware for Wix-managed headless projects.

Independently verify AxioRank audit receipts, agent-commerce conformance seals, agent reputation credentials, and coding-session seals offline. Merkle inclusion proofs, Ed25519 signed tree heads, delegation provenance, human-approval signatures, witness c

The security gateway for AI agents. One local, OpenAI-compatible endpoint for every model provider, with guardrails on by default and a signed, offline-verifiable receipt on every response.

Scan any MCP server for prompt injection, tool poisoning, leaked secrets, and destructive capabilities. One command, no key, no signup.

AxioRank agent red-team corpus: a versioned library of attack scenarios (secrets, injection, destructive ops, kill chains) plus benign controls. Pure data, zero runtime dependencies.

Govern AI coding agents (Claude Code, Cursor, Codex, GitHub Copilot CLI) from a hook. It blocks dangerous tool calls locally and offline (destructive commands, secret exfiltration, prompt-injected tool results), then optionally reports the session to Axio