TealTiger
Deterministic security, cost governance, and audit evidence for AI applications.
- Source repo
- agentguard-ai/tealtiger
- Stars
- ★ 37
- Last updated
- 1mo ago
- License
- Apache-2.0
- Primary language
- TypeScript
- FA score
- 48/100 · Major gaps
At a glance
- How it runs
- Works with
- Universal · cross-platformOpenAI API · Claude API
- Cost
- Free software; you pay for model usage
- Setup effort
- Low · running in minutes
- You'll need
- Typical use
- Engineering teams maintaining OpenAI or Anthropic applications that want request auditing, cost attribution, and PII reporting without redesigning their call flow.
- Not a fit if
- Enterprises needing production multi-tenancy, RBAC, or SSO today
- Teams requiring verified observe() coverage for Cohere or Mistral
- Teams expecting every SDK source tree directly inside one repository
- Website
- tealtiger.ai ↗ set on GitHub
- Maintenance
- Commits in the last 6 monthslast commit 2026-10-09latest release tealtiger-mcp-v1.1.0 (2026-10-05)
- Source review
- 48/100 · Major gaps
What does this agent do, and when should you use it?
TealTiger is an open-source runtime governance SDK for LLM applications and AI agents, with TypeScript and Python interfaces. It wraps model clients to inspect requests for prompt injection, PII, secrets, and moderated content while tracking costs and budgets. TealEngine evaluates deterministic policies in ENFORCE, MONITOR, or REPORT_ONLY modes and can produce a structured record for every verdict. Named modules such as TealProof, TealAudit, TealMonitor, TealMemory, and TealDrift cover evidence, auditing, cost, memory, and behavioral governance. The core runs as an embedded library without separate governance infrastructure, while an optional dashboard can be run locally; this repository is a hub, with the primary SDK sources stored in separate repositories or Git submodules.
An application can wrap an existing LLM client with observe(client) to collect per-request costs, correlation IDs, audit events, REPORT_ONLY PII findings, and P50/P95/P99 behavioral baselines; freeze() and unfreeze() control its emergency stop state. For active blocking, TealOpenAI accepts options such as piiDetection, promptInjection, contentModeration, per-request limits, and daily budgets. TealEngine reads policies and returns deterministic ALLOW or denial decisions, while related modules govern tool allowlists, session timing, memory-write provenance, and inherited workflows. Decisions can yield reconstructable evidence exported as JSON, SARIF, or JUnit XML; TealProof also describes Merkle-tree receipts and RFC 3161 timestamping. The dashboard fetches panels independently and displays request and cost KPIs, budget forecasts, defense stages, drift alerts, freeze state, model routing, and protocol-policy results.
- Engineering teams maintaining OpenAI or Anthropic applications that want request auditing, cost attribution, and PII reporting without redesigning their call flow.
- Teams operating tool-using agents that need tool allowlists and argument-level PII rules to stop unauthorized or sensitive operations.
- Security platforms introducing deterministic policies gradually through ENFORCE, MONITOR, and REPORT_ONLY modes while retaining reconstructable evidence.
- Teams controlling multi-model expenditure through per-request, session, or daily ceilings and circuit breakers for runaway loops.
- Operators managing a fleet of agents who need a local dashboard for spend, denials, behavioral drift, and frozen-agent status.
Key setup facts
Pulled from the install and usage notes below. The project README has the current values.
- Install / run
pip install tealtigernpm install tealtigernpx ts-node scripts/validate-policy.ts ./my-policy.json
- Environment variables
OPENAI_API_KEY- Config files
./my-policy.json
How do you install or deploy this agent?
Choose either supported SDK path. For Python:
pip install tealtigerFor a TypeScript project:
npm install tealtigerThe supplied governance tutorial can run without a model API key:
python examples/quickstart_governance.pyThe live OpenAI examples require OPENAI_API_KEY in the environment. This hub does not contain all SDK source directly; clone its submodules when a complete source checkout is required:
git clone --recurse-submodules https://github.com/agentguard-ai/tealtiger.gitHow do you use this agent?
The minimal Python flow creates a guarded TealOpenAI client and makes a normal chat request:
import os
from tealtiger import TealOpenAI
client = TealOpenAI(
api_key=os.environ["OPENAI_API_KEY"],
guardrails={"prompt_injection": True},
)
response = client.chat.completions.create(
model="gpt-4o-mini",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.security.decision)For lower-touch observation, wrap an existing client:
from tealtiger.observe import observe, freeze
client = observe(OpenAI())
print(client.get_cost())Run the local dashboard API and web interface separately:
cd dashboard/api
npm run devcd dashboard/web
npm run devThe API uses port 3100 and the UI uses port 3000. Validate a JSON policy before CI/CD or runtime use with:
npm install
npx ts-node scripts/validate-policy.ts ./my-policy.jsonWhat are this agent's strengths and limitations?
- The governance path does not call an LLM, so identical input and policy are intended to produce reproducible decisions.
observe(client)combines cost records, correlation IDs, PII reporting, behavioral baselines, and emergency freezing behind a small integration surface.- Both TypeScript and Python SDKs are documented, with provider paths for OpenAI, Anthropic, Gemini, Bedrock, Azure OpenAI, Cohere, and Mistral.
- Evidence formats include JSON, SARIF, JUnit XML, OpenTelemetry spans, and described SIEM integration points.
- Governance adapters are documented for 13 agent frameworks, including concrete packages for LangChain, Vercel AI SDK, PydanticAI, Haystack, and CAMEL-AI.
- Provider maturity varies: only selected OpenAI and Anthropic models are marked as end-to-end verified through
observe(); Gemini, Bedrock, and Azure OpenAI lack that integration, while Cohere and Mistral lack SDK hooks. - The repository is a hub rather than the complete SDK source tree; the main TypeScript and Python implementations live in separate repositories.
- Multi-tenancy, RBAC, SSO, policy staging, and scheduled compliance reports are listed as planned v1.5.0 work.
- The SaaS control plane, Kubernetes Operator, shadow-agent discovery, and remote kill switch are future v2.0.0 items rather than current capabilities.
- Live model calls still require provider credentials, network access, and any usage fees charged by those providers.
How does this agent compare with similar options?
Unlike probabilistic safety filters, TealTiger explicitly uses a deterministic, no-LLM policy path: the same input and policy should yield the same outcome, with reconstructable evidence for each verdict. The source does not provide a feature-by-feature comparison against a named competing product.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| TealTiger This agent | 48 · Major gaps | Library / SDKFree + model costs | ★ 37 | 1mo ago | TypeScript | OpenAI API · Claude API |
| Node9 | 68 · Some gaps | CLIFreemium | ★ 225 | today | TypeScript | Codex · Claude Code |
| OpenBot: AI Coworkers You Can Trust with Real Work | 87 · Good | Self-hosted serviceFree + model costs | ★ 13k | today | TypeScript | — |
| Kontext | 80 · Good | CLIFreemium | ★ 235 | today | Go | Codex · Claude Code |
How does FollowAgents rate this agent?
Why each dimension lost points
The README and SECURITY.md advise keeping keys in environment variables or secret managers and describe PII, budgets, audit, and policy controls. The SDK implementation is outside this packet, so actual permission scope, data flows, confirmation behavior, and side effects cannot be established. observe instrumentation and freeze/unfreeze are described without rollback details. The README, security policy, workflows, and contact channels provide partial project attribution, while publisher identity remains unverified.
The docs describe deterministic policy evaluation, fault isolation, and error handling, and list test entry points for APIs, streams, policy validation, and adapter coverage. However, provider counts differ across README sections (7, 12, and 13), and the repository calls itself a hub whose SDK implementations are in external repositories; this weakens consistency and dependency-availability evidence. Troubleshooting maps common errors to causes and fixes.
The material targets developers building LLM applications and covers OpenAI, multiple providers, framework adapters, and a progression from observation to enforcement. It gives some security-model boundaries, but does not fully clarify which capabilities are in this hub, when they trigger automatically, or the precise boundaries of the external SDKs.
The README organizes quick starts, features, providers, documentation, and roadmap, and links FAQ, troubleshooting, and security policy. It documents npm and pip installation and includes the full Apache 2.0 license. Version information and plans are extensive, but release and roadmap claims are not backed by a conventional changelog in the supplied material; SECURITY.md clearly lists supported versions, reporting channels, and response targets.
Examples show usable structured outputs for policy decisions, governance events, cost tracking, and guardrails, and describe meaningful governance features for agent applications. The core SDK source is absent from this packet, so feature coverage cannot be confirmed. Provider charges and deployment or configuration costs are not compared, limiting cost-benefit assessment.
The README, FAQ, security policy, quick start, and troubleshooting provide traceable explanations of setup, policy modes, errors, and the security model. Most capability claims remain project assertions in this sample, which lacks SDK implementation or independent corroboration. The docs do not consistently distinguish design claims, example behavior, and production results, and provider counts conflict.
- This repository points to external TypeScript and Python SDK source repositories; those implementations were not reviewed, and no code or tests were run.
- Provider counts conflict within the README. Before production use, verify actual feature support, data handling, and failure behavior in the intended SDK version.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
tealtiger-mcp-v1.1.0 tealtiger-mcp v1.1.0
Release notes excerpt
tealtiger-mcp v1.1.0 MCP server exposing TealTiger guardrails, cost estimation and budget checks as tools for Claude and other MCP clients. Install ```bash pip install tealtiger-mcp==1.1.0 ``` Links - [PyPI](https://pypi.org/project/tealtiger-mcp/1.1.0/) - [Changelog](https://github.com/agentguard-ai/tealtiger/blob/main/packages/tealtiger-mcp/CHANGELOG.md) - [Documentation](https://docs.tealtiger.ai)View on GitHub ↗ -
v1.4.0 ## What's New in v1.4.0
Release notes excerpt
observe() — Zero-Config Instrumentation One line adds cost tracking, audit logging, PII detection, and behavioral baselines to any LLM client. Key Features - observe(client) — Zero-config proxy wrapping for 12 LLM providers - Automatic Cost Tracking — Per-request, per-session, per-agent cost accumulation - Behavioral Baseline — P50/P95/P99 latency and token distribution profiling - PII Detection (REPORT_ONLY) — Passive scanning without blocking - freeze() / unfreeze() — Instant kill switch for any agent …View on GitHub ↗ -
langchain-tealtiger-v0.1.0 langchain-tealtiger v0.1.0
Release notes excerpt
langchain-tealtiger v0.1.0 Deterministic governance middleware for LangChain agents. Install ```bash pip install langchain-tealtiger==0.1.0 ``` Quick Start ```python from langchain.agents import create_agent from langchain_tealtiger import TealTigerMiddleware agent = create_agent( model="claude-sonnet-4-6", tools=[...], middleware=[ TealTigerMiddleware( policies=[ {"type": "tool_allowlist", "tools": ["search", "calculator"]}, {"type": "cost_limit", "max_per_session": 5.00}, ], …View on GitHub ↗
FAQ
Do I have to use OpenAI?
observe().Can I evaluate it without a model API key?
pip install tealtiger, the five-minute examples/quickstart_governance.py tutorial is documented as requiring no API key. Real hosted-model requests still require that provider's credentials.Does every policy violation block the request?
Does it require a separate governance service?
How can an operator stop a runaway agent?
freeze() and unfreeze() provide an immediate kill-switch mechanism. TealEngine FREEZE rules additionally describe immutable emergency controls with tamper detection.Is TealTiger free?
Is TealTiger open source?
Is TealTiger safe to use?
Is TealTiger still maintained?
Who makes TealTiger?
What is TealTiger's official website?
Thanks — we use this to improve the page.