Knowledge base

Documentation

Everything you need to deploy, configure, and extend the AI-FW gateway, guides, hands-on tutorials, and API references.

Getting Started

Start here: what AI-FW is and how to get your gateway running in minutes.

Guides

Deep dives into each capability: guardrails, routing, semantic analysis, identity, risk, and reliability.

01

Prompt & response guardrails

How AI-FW inspects every prompt and response, built-in rules, custom rules, evaluation order, and rule actions.

02

Model routing & registry

How AI-FW decides which model handles a request, the model inventory, default model and backend, strict mode, and conditional routing rules.

03

Semantic intent analysis

Judge the meaning of AI traffic, not just its text, natural-language policies, block and flag thresholds, and three scoring engines.

04

Identity & access

Authentication modes, API keys, key precedence, streaming modes, mTLS, Kerberos, OIDC SSO, SCIM, and admin roles.

05

Risk profiles & auto-block

Rolling risk scores for users, agents, and IPs, with automatic blocking, category risk cards, and a metadata-only audit trail.

06

Reliability & caching

Per-model resilience (retries, failover, distribution) and the opt-in completion cache, exact and semantic, with strict tenant isolation.

07

Prompt compression

Cut upstream token spend with semantic-gated prompt compression, rule, aggressive, and LLM tiers that never change the meaning.

08

Claude inference hooks

Act as the AI security server for Anthropic's Inference Hooks - verified, scanned, and machine-actionable verdicts before inference proceeds.

09

Observability & dashboard

The AI-FW dashboard - traffic stats, latency, guardrail snapshot, 24-hour metrics, and the live Activity view.

10

Audit logs & export

The metadata-only audit trail - logging policy, retention, audit events, and exporting to your SIEM via pull API or syslog.

11

Admin-issued API keys

Issue API keys for agent and tool access - hashed storage, immediate revocation, and per-key identity in the audit trail.

Tutorials

Hands-on walkthroughs that connect real clients to the gateway.

How-To

Focused recipes for common admin tasks.

API Reference

The OpenAI-compatible, Anthropic, and agent-protocol endpoints.

Admin Reference

Every admin page and its options: Settings, Model Inventory, Rules Manager, Risk Profiles, Audit, and more.

01

Settings reference

Every tab of the Settings page - General, Access, Identity, Model, Routing, Guardrails & Inspection, Performance, Audit, Export, Integrations, and Licensing.

02

Model Inventory reference

The Model Inventory page and every option: filters, enable/disable, add/edit/delete, protocols, resilience, compression, advanced parameters, groups, and routing rules.

03

Rules Manager reference

Built-in rules, custom rules, rule types and actions, per-rule options, priority, and AI-assisted rule creation.

04

Risk Profiles reference

User, agent, and IP risk leaderboards, category risk cards, risk auto-block, and reset actions.

05

Audit Logs reference

The transaction log, filters, the Events tab, and page-size settings.

06

Dashboard reference

Every card and control on the AI-FW Dashboard - traffic stats, latency, guardrail snapshot, 24-hour metrics, and charts.

07

M365 Copilot admin reference

The M365 Copilot page - agent catalog, agent registry, usage analytics, and interaction export with app-only authentication.

08

Users reference

Local users, roles, passwords, SSO, SCIM provisioning, and user groups.

09

Agent API keys reference

The Agent API Keys page - add, generate, label, and revoke keys with hashed storage and per-key identity.

10

Agents reference

The agent registry admin page - onboarding modes, identity proof, approval, groups and tags, webhooks, and deboarding.

11

CA integrations reference

SCEP and ACME enrollment, the central CA trust store, and server-certificate enrollment.

12

Agent Trust reference

The Agent Trust dashboard - agent inventory, approvals, tasks, and trust settings.

Community Edition

The free edition license, the published monthly token allowance, and how the gateway enforces it.