Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.tealtiger.ai/llms.txt

Use this file to discover all available pages before exploring further.

TealTiger is a runtime AI governance framework for LLM and agentic applications. It provides deterministic policy enforcement across security, cost, reliability, compliance, and evidence — embedded directly into application code, with no infrastructure or SaaS dependencies. These docs are written for developers, architects, security teams, and governance leads building or operating production AI systems.

​
Why TealTiger

TealTiger is not a guardrail library. It is a governance framework that grows with your AI system — from security basics to full enterprise governance.

Governance Framework

Six governance dimensions in Bundle 1.1: Security, Cost, Authority, Reliability, Evidence, and Compliance. More dimensions ship with each bundle.

Zero Infrastructure

SDK-only architecture. No servers, control planes, or external services. Embeds directly into your application runtime.

Deterministic Decisions

Same input + same policy = same decision. Typed Decision contracts with explicit actions, reason codes, and risk scores.

Audit-Grade Evidence

Every enforcement decision produces versioned audit events with redaction-by-default, correlation IDs, and traceability.

​
Get Started

If you are new to TealTiger, start here:

Quickstart

Get TealTiger running in under five minutes.

Your First Policy

Write, test, and deploy your first policy.

Core Concepts

Learn how decisions are evaluated and enforced.

API Reference

Explore the full TypeScript and Python SDK APIs.

​
Key Capabilities

​
Governance Dimensions (Bundle 1.1)

TealTiger governs six dimensions in the current release:
  • Security — Prompt injection, PII detection, content moderation, code execution policies
  • Cost — Budget limits, rate limiting, anomaly detection
  • Authority — Tool-level allow/deny, identity-based access control
  • Reliability — Circuit breaker, failure thresholds, recovery
  • Evidence — Versioned audit events, redaction-by-default, correlation IDs
  • Compliance — Policy rollout modes, deterministic Decision contract, testing framework

​
Policy Rollout Modes

Safely introduce governance controls using three rollout modes:
  • ENFORCE — Block actions that violate policy
  • MONITOR — Record violations without blocking
  • REPORT_ONLY — Collect data for tuning and analysis

​
Developer Experience

  • TypeScript and Python SDKs
  • Policy testing framework and CLI
  • CI/CD-friendly workflows
  • No telemetry sent to external services

​
Supported Providers

TealTiger is provider-agnostic and works across major LLM platforms:
  • ✅ OpenAI (GPT-4, GPT-3.5)
  • ✅ Anthropic (Claude)
  • ✅ Google Gemini
  • ✅ AWS Bedrock (Claude, Titan, Jurassic, Command, Llama)
  • ✅ Azure OpenAI
  • ✅ Cohere (Chat, RAG, Embeddings)
  • ✅ Mistral AI

​
Open Source

TealTiger SDKs are open source under the MIT license and will remain free to use. A future SaaS offering, if introduced, will be provided as a separate commercial product.

View on GitHub

Star the project, explore the codebase, and contribute.

​
Community and Support

Documentation

Guides, concepts, and API references.

GitHub Issues

Report bugs and request features.

Email Support

Contact the TealTiger team.
Current Version: v1.1.0
License: MIT
Status: Production Ready