Skip to content

Same agent Smaller bill. The tools and infrastructure to get more work from your AI. Fewer tokens. Less compute. Lower cost.

Align README, docs, and package metadata to agent harness engineering (#1685)

Aman Agarwal897838a26e44
diff --git a/README.md b/README.md
index d2498d4e..750fd286 100644
--- a/README.md
+++ b/README.md
@@ -1,13 +1,14 @@
<div align="center">
-<img src="https://github.com/openlit/.github/blob/main/profile/assets/wide-logo-no-bg.png?raw=true" alt="OpenLIT Logo" width="30%">
+<img src="https://github.com/openlit/.github/blob/main/profile/assets/wide-logo-no-bg.png?raw=true" alt="OpenLIT logo" width="30%">
-# Open-source observability & evaluation for AI agents
+# OpenLIT: the open-source agent harness engineering platform
-**Trace, evaluate, debug, and optimize AI applications and coding agents with OpenTelemetry.**
+**Trace, evaluate, guard, and improve AI agents and coding agents with OpenTelemetry.**
+Agent observability · Evals · Guardrails · Prompt management · Cost & GPU monitoring · Self-host free (Apache 2.0)
-<a href="https://docs.openlit.io/latest/features/agent-observability">
- <img src="docs/images/readme-hero-coding-agent-trace.png" alt="OpenLIT coding agent trace view" width="90%">
+<a href="https://docs.openlit.io/latest/openlit/observability/agents/overview">
+ <img src="docs/images/readme-hero-coding-agent-trace.png" alt="OpenLIT agent harness trace view showing LLM calls, tool calls and cost" width="90%">
</a>
<p>
@@ -22,15 +23,28 @@
[![Slack](https://img.shields.io/badge/Slack-join-purple?logo=slack)](https://join.slack.com/t/openlit/shared_invite/zt-2etnfttwg-TjP_7BZXfYg84oAukY8QRQ)
[![X](https://img.shields.io/badge/follow-%40openlit__io-black?logo=x)](https://twitter.com/openlit_io)
-**[Documentation](https://docs.openlit.io/) · [Quickstart](https://docs.openlit.io/) · [Examples](https://github.com/openlit/openlit/tree/main/examples) · [Join Slack](https://join.slack.com/t/openlit/shared_invite/zt-2etnfttwg-TjP_7BZXfYg84oAukY8QRQ)**
+**[Website](https://openlit.io) · [Documentation](https://docs.openlit.io/latest/overview) · [Quickstart](#-get-started-in-5-minutes) · [Compare](https://openlit.io/compare) · [Join Slack](https://join.slack.com/t/openlit/shared_invite/zt-2etnfttwg-TjP_7BZXfYg84oAukY8QRQ)**
</div>
---
-## See what your AI agents are actually doing
+## What is OpenLIT?
-AI applications are no longer just LLM calls.
+**OpenLIT is an open-source agent harness engineering platform.** It gives teams OpenTelemetry-native tracing, evaluations, guardrails, prompt and context management, and cost and GPU monitoring for the *harness* around their AI agents and coding agents, so every agent failure can be traced, scored, and turned into a harness fix. OpenLIT is free to self-host under Apache 2.0 and works with any model, framework, or harness: Claude Code, Codex, Cursor, OpenAI Agents SDK, LangGraph, CrewAI, and 70+ more integrations.
+
+### What is agent harness engineering?
+
+An AI agent is a **model plus a harness**. The harness is everything except the model: tools, context, prompts, memory, hooks, guardrails, and feedback loops. **Agent harness engineering** is the discipline of designing, measuring, and improving that harness so agents are reliable in production. The core loop is *run → observe → evaluate → fix the harness → verify*, and OpenLIT gives you each step:
+
+| Harness engineering step | OpenLIT feature |
+|---|---|
+| **Observe** every LLM call, tool call, MCP request, retrieval and agent step | Agent observability & OpenTelemetry LLM tracing |
+| **Evaluate** quality, safety and cost on real traces | Evals (LLM-as-a-judge, programmatic, human feedback) + CI gates |
+| **Guard** the agent at runtime | Guardrails (prompt injection, sensitive topics, topic restriction) |
+| **Fix** the harness without redeploying | Prompt Hub, Context, Rule Engine, Vault (secrets) |
+| **Compare** models and prompts before shipping | OpenGround |
+| **Account** for cost, latency and hardware | Cost tracking, custom pricing, GPU monitoring (NVIDIA, AMD, Intel) |
A production agent can involve:
@@ -53,10 +67,6 @@ flowchart TD
style O fill:#F97316,stroke:#7C2D12,color:#fff
```
-**OpenLIT gives you visibility across the entire workflow.**
-
-Trace every LLM call, tool invocation, prompt, agent step, token, cost, error, and evaluation — using OpenTelemetry.
-
---
## ⚡ Get started in 5 minutes
@@ -126,7 +136,7 @@ Open your dashboard and start exploring your AI application's traces, metrics, c
---
-# 🤖 Observe Claude Code, Cursor & Codex
+# 🤖 Coding agent observability: Claude Code, Cursor, Codex & Windsurf
AI coding agents are powerful — but understanding what they actually did can be difficult.
@@ -196,7 +206,7 @@ Explore the resulting sessions in the **Coding Agents** dashboard.
---
-# 🔍 What OpenLIT gives you
+# 🔍 Agent observability & LLM tracing (OpenTelemetry)
## Traces
@@ -210,7 +220,7 @@ All represented using OpenTelemetry.
---
-## 💰 AI cost observability
+## 💰 LLM cost tracking
Track the cost of your AI applications across:
@@ -222,7 +232,7 @@ Support custom pricing for custom and fine-tuned models.
---
-## 🧪 AI evaluations
+## 🧪 Agent evals: LLM-as-a-judge, programmatic evals & CI gates
Automatically evaluate LLM and agent outputs using LLM-as-a-Judge evaluations.
@@ -242,7 +252,7 @@ to:
---
-## 🐛 Debug production AI
+## 🐛 Debug production AI agents
Find the requests that matter.
@@ -303,7 +313,7 @@ flowchart TD
---
-# 🧠 Prompt management
+# 🧠 Prompt management (Prompt Hub) & context
Use **Prompt Hub** to:
@@ -321,6 +331,8 @@ prompt = openlit.prompts.get(
Keep prompt management separate from application code while maintaining version control and observability.
+Use **Context** to store reusable RAG content once and have the Rule Engine return the right piece at runtime. See the [Context docs](https://docs.openlit.io/latest/openlit/prompts-experiments/context/overview).
+
---
# ⚙️ Rule Engine
@@ -348,9 +360,39 @@ THEN
---
-# 🔌 OpenTelemetry-native
+# 🛡️ Guardrails: prompt injection & sensitive-topic detection
+
+Use OpenLIT SDK guardrails to detect and block risky prompts at runtime:
+
+- Prompt injection / jailbreak attempts
+- Sensitive topics
+- Topic restriction
+
+Guardrail checks are traced with OpenTelemetry so you can see when and why a request was blocked. See the [guardrails docs](https://docs.openlit.io/latest/sdk/features/guardrails) and [quickstart](https://docs.openlit.io/latest/openlit/quickstart-guard).
+
+---
+
+# 🔐 Vault: LLM API key management
+
+Store LLM API keys and other secrets in **Vault**, then retrieve them at runtime via the SDK or API — without hard-coding credentials in application code. See the [Vault docs](https://docs.openlit.io/latest/openlit/developer-resources/vault/overview).
+
+---
+
+# 🆚 OpenGround: compare models side by side
+
+Run the same prompt across multiple providers in **OpenGround** and compare cost, latency, and output quality before you ship. See the [OpenGround docs](https://docs.openlit.io/latest/openlit/prompts-experiments/openground/overview).
+
+---
+
+# 🖥️ GPU monitoring for LLM inference
+
+Monitor NVIDIA, AMD, and Intel GPUs used for LLM inference with the OpenTelemetry GPU collector: utilization, memory, power, and temperature, correlated with your traces. See the [GPU collector docs](https://docs.openlit.io/latest/gpu-collector/overview).
+
+---
+
+# 🔌 OpenTelemetry-native (GenAI semantic conventions)
-OpenLIT is built around **OpenTelemetry**, rather than creating a proprietary telemetry format.
+OpenLIT is built around **OpenTelemetry**, rather than creating a proprietary telemetry format. Traces and metrics follow OpenTelemetry GenAI semantic conventions (`gen_ai.*`).
Your telemetry can flow through the OpenTelemetry ecosystem:
@@ -491,6 +533,54 @@ The goal is simple:
---
+## ❓ FAQ
+
+**What is OpenLIT?**
+OpenLIT is an open-source agent harness engineering platform. It provides OpenTelemetry-native tracing, evaluations, guardrails, prompt management, and cost and GPU monitoring for AI agents and coding agents, and it is free to self-host under Apache 2.0.
+
+**What is an agent harness?**
+An agent harness is everything in an AI agent except the model: the tools, context, prompts, memory, hooks, guardrails, and feedback loops that turn a model into a working agent. Claude Code, Codex, and frameworks like LangGraph or CrewAI are harnesses.
+
+**What is agent harness engineering?**
+Agent harness engineering is the discipline of designing, measuring, and improving the harness around a model so agents are reliable in production. Teams run agents on real tasks, observe failures in traces, evaluate them, fix the harness (prompts, tools, rules, guardrails), and verify the fix with regression evals.
+
+**Is OpenLIT an agent framework or a harness runtime?**
+No. OpenLIT doesn't run your agent loop. It instruments and improves whatever harness you already use, through OpenTelemetry, so you can switch models, frameworks, or harnesses without losing your traces, evals, or prompts.
+
+**How is OpenLIT different from Langfuse, LangSmith, or Arize Phoenix?**
+OpenLIT is Apache-2.0 and OpenTelemetry-native end to end, adds GPU monitoring, guardrails, Vault, and coding-agent observability in the same self-hosted platform, and exports to any OTLP backend. See https://openlit.io/compare for feature-by-feature comparisons.
+
+**Does OpenLIT work with Claude Code, Codex, and Cursor?**
+Yes. The `openlit` CLI ingests each coding agent's hook events and maps them to OpenTelemetry `gen_ai.*` conventions, with no SDK and no code changes.
+
+**Can I run agent evals in CI?**
+Yes. Run LLM-as-a-judge and programmatic evals online on production traces, or offline through the SDK as CI/CD gates.
+
+**Is OpenLIT free?**
+Yes. Self-hosted OpenLIT is free under Apache 2.0, with no license key and no per-trace fee.
+
+**Does OpenLIT add latency?**
+No proxy is required. SDKs instrument in-process and export telemetry asynchronously over OTLP.
+
+**Can I send data to Grafana, Datadog, or my existing OpenTelemetry Collector?**
+Yes. OpenLIT emits standard OTLP traces and metrics, so they can go to OpenLIT, Grafana, Datadog, New Relic, SigNoz, or any OTLP backend.
+
+---
+
+## 🆚 How OpenLIT compares
+
+Feature-by-feature comparisons on [openlit.io/compare](https://openlit.io/compare):
+
+- [OpenLIT vs Langfuse](https://openlit.io/compare/openlit-vs-langfuse)
+- [OpenLIT vs LangSmith](https://openlit.io/compare/openlit-vs-langsmith)
+- [OpenLIT vs Arize Phoenix](https://openlit.io/compare/openlit-vs-arize-phoenix)
+- [OpenLIT vs Braintrust](https://openlit.io/compare/openlit-vs-braintrust)
+- [OpenLIT vs Helicone](https://openlit.io/compare/openlit-vs-helicone)
+- [OpenLIT vs Datadog](https://openlit.io/compare/openlit-vs-datadog)
+- [OpenLIT vs Comet Opik](https://openlit.io/compare/openlit-vs-comet-opik)
+
+---
+
# 🌎 Community
OpenLIT is open source and built with the AI engineering community.
@@ -582,7 +672,7 @@ Names for the same contributors are on [openlit.io/about-us](https://openlit.io/
<div align="center">
-### Build AI systems you can actually understand.
+### Build agent harnesses you can actually measure and improve.
**⭐ Star OpenLIT on GitHub**
diff --git a/docs/docs.json b/docs/docs.json
index 1b60d869..8aff6277 100644
--- a/docs/docs.json
+++ b/docs/docs.json
@@ -2,7 +2,7 @@
"$schema": "https://mintlify.com/docs.json",
"theme": "aspen",
"name": "OpenLIT",
- "description": "OpenLIT is the leading open-source AI observability platform with zero-code instrumentation for LLMs, vector databases, and AI frameworks. Monitor OpenAI, Anthropic, LangChain, LlamaIndex with OpenTelemetry. Features include cost tracking, performance metrics, prompt management, and enterprise-grade security for production AI applications.",
+ "description": "OpenLIT is the open-source agent harness engineering platform: OpenTelemetry-native agent observability, evals, guardrails, prompt management, and cost and GPU monitoring for AI agents and coding agents. Free to self-host under Apache 2.0.",
"colors": {
"primary": "#F36C06",
"light": "#FF8A3D",
@@ -81,6 +81,15 @@
}
]
},
+ {
+ "group": "Concepts",
+ "pages": [
+ "latest/concepts/glossary",
+ "latest/glossary/what-is-ai-observability",
+ "latest/glossary/what-is-ai-evaluation",
+ "latest/glossary/what-is-prompt-engineering"
+ ]
+ },
{
"group": "Observability",
"pages": [
@@ -161,7 +170,7 @@
"pages": [
"latest/openlit/dashboards/widgets/overview",
"latest/openlit/dashboards/widgets/area-chart-widget",
- "latest/openlit/dashboards/widgets/bar-chart-widget",
+ "latest/openlit/dashboards/widgets/bar-chart-widget",
"latest/openlit/dashboards/widgets/line-chart-widget",
"latest/openlit/dashboards/widgets/pie-chart-widget",
"latest/openlit/dashboards/widgets/stat-widget",
@@ -578,9 +587,65 @@
}
},
"contextual": {
- "options": ["copy", "chatgpt", "claude"]
+ "options": [
+ "copy",
+ "chatgpt",
+ "claude"
+ ]
},
"redirects": [
+ {
+ "source": "/latest/features/evaluations",
+ "destination": "/latest/openlit/evaluations/overview"
+ },
+ {
+ "source": "/latest/features/guardrails",
+ "destination": "/latest/sdk/features/guardrails"
+ },
+ {
+ "source": "/latest/features/gpu",
+ "destination": "/latest/gpu-collector/overview"
+ },
+ {
+ "source": "/latest/features/agent-observability",
+ "destination": "/latest/openlit/observability/agents/overview"
+ },
+ {
+ "source": "/latest/quickstart",
+ "destination": "/latest/openlit/quickstart-ai-observability"
+ },
+ {
+ "source": "/latest/destinations/intro",
+ "destination": "/latest/sdk/destinations/overview"
+ },
+ {
+ "source": "/latest/integrations/nvidia-gpu",
+ "destination": "/latest/gpu-collector/gpus/nvidia"
+ },
+ {
+ "source": "/latest/integrations/amd-gpu",
+ "destination": "/latest/gpu-collector/gpus/amd"
+ },
+ {
+ "source": "/latest/integrations/intel-gpu",
+ "destination": "/latest/gpu-collector/gpus/intel"
+ },
+ {
+ "source": "/latest/integrations/embedchain",
+ "destination": "/latest/sdk/integrations/overview"
+ },
+ {
+ "source": "/latest/sdk/integrations/embedchain",
+ "destination": "/latest/sdk/integrations/overview"
+ },
+ {
+ "source": "/latest/integrations/:slug",
+ "destination": "/latest/sdk/integrations/:slug"
+ },
+ {
+ "source": "/latest/connections/:slug",
+ "destination": "/latest/sdk/destinations/:slug"
+ },
{
"source": "/latest/openlit/observability/tracing",
"destination": "/latest/openlit/observability/telemetry/traces"
diff --git a/docs/latest/concepts/glossary.mdx b/docs/latest/concepts/glossary.mdx
new file mode 100644
index 00000000..f5d8936d
--- /dev/null
+++ b/docs/latest/concepts/glossary.mdx
@@ -0,0 +1,69 @@
+---
+title: "Agent Harness Engineering Glossary"
+sidebarTitle: Glossary
+description: "Short definitions of agent harness, harness engineering, agent observability, agent evals, guardrails, and trajectory and tool-call evaluation."
+---
+
+This glossary defines the core terms OpenLIT uses for **agent harness engineering**. Each entry opens with a one-sentence definition you can quote.
+
+## Agent harness
+
+An **agent harness** is everything in an AI agent except the model: tools, context, prompts, memory, hooks, guardrails, and feedback loops that turn a model into a working agent. Claude Code, Codex, and frameworks such as LangGraph or CrewAI are harnesses. OpenLIT does not replace your harness; it observes, evaluates, and improves it via OpenTelemetry.
+
+Related: [AI agent observability](/latest/openlit/observability/agents/overview) · [Coding agent observability](/latest/openlit/coding-agents/overview)
+
+## Agent harness engineering
+
+**Agent harness engineering** (or harness engineering) is the discipline of designing, measuring, and improving the harness around a model so agents are reliable in production. The core loop is *run → observe → evaluate → fix the harness → verify*. OpenLIT is the open-source platform for that loop: tracing, evals, guardrails, prompt and context management, and cost and GPU monitoring.
+
+Related: [What is OpenLIT?](/latest/overview)
+
+## Agent observability
+
+**Agent observability** is the practice of tracing every LLM call, tool call, MCP request, retrieval, and agent step so you can see cost, latency, errors, and behavior across the full harness — not only a single model response. OpenLIT provides OpenTelemetry-native agent observability for application agents and coding agents.
+
+Related: [Agents](/latest/openlit/observability/agents/overview) · [Telemetry](/latest/openlit/observability/telemetry/overview) · [What is AI Observability?](/latest/glossary/what-is-ai-observability)
+
+## Agent evals
+
+**Agent evals** (agent evaluation) score agent and LLM outputs for quality, safety, and task success. In OpenLIT you can run LLM-as-a-judge and programmatic evals on production traces, collect human feedback, and use the same criteria as CI/CD regression gates.
+
+Related: [Evaluations](/latest/openlit/evaluations/overview) · [LLM-as-a-judge](/latest/openlit/evaluations/llm-as-a-judge) · [Programmatic evals](/latest/openlit/evaluations/programmatic-evals) · [What is AI Evaluation?](/latest/glossary/what-is-ai-evaluation)
+
+## Guardrails
+
+**Guardrails** are runtime checks that block or flag risky agent behavior before it reaches users or downstream systems — for example prompt injection, sensitive topics, and topic restriction. OpenLIT SDK guardrails emit OpenTelemetry traces so blocked requests stay visible in the same observability loop.
+
+Related: [Guardrails](/latest/sdk/features/guardrails) · [Guardrails quickstart](/latest/openlit/quickstart-guard)
+
+## Trajectory evaluation / tool-call evaluation
+
+**Trajectory evaluation** scores the path an agent took (which tools it called, in what order, and with what outcomes), not only the final answer. **Tool-call evaluation** focuses on whether individual tool invocations were correct, necessary, and successful. Both help turn a failed run into a harness fix (better tools, prompts, or rules) rather than a one-off prompt tweak.
+
+Related: [Agent evals](/latest/openlit/evaluations/overview) · [Agent observability](/latest/openlit/observability/agents/overview)
+
+## Prompt management
+
+**Prompt management** is versioning, publishing, and fetching prompts at runtime so you can fix the harness without redeploying application code. OpenLIT Prompt Hub links prompt versions to traces and evaluations.
+
+Related: [Prompt Hub](/latest/openlit/prompts-experiments/prompt-hub/overview) · [What is Prompt Engineering?](/latest/glossary/what-is-prompt-engineering)
+
+## OpenTelemetry GenAI semantic conventions
+
+**OpenTelemetry GenAI semantic conventions** are the standard attribute and span names (`gen_ai.*`) for LLM and agent telemetry. OpenLIT follows these conventions so traces export cleanly to OpenLIT or any OTLP backend.
+
+Related: [SDK overview](/latest/sdk/overview) · [Destinations](/latest/sdk/destinations/overview)
+
+---
+
+<CardGroup cols={3}>
+ <Card title="What is OpenLIT?" href="/latest/overview" icon="book-open">
+ Platform overview and the harness engineering loop
+ </Card>
+ <Card title="Agent observability" href="/latest/openlit/observability/agents/overview" icon="radar">
+ Trace every agent step from OpenTelemetry data
+ </Card>
+ <Card title="Agent evals" href="/latest/openlit/evaluations/overview" icon="clipboard-check">
+ Score production traces and gate releases in CI
+ </Card>
+</CardGroup>
diff --git a/docs/latest/gpu-collector/overview.mdx b/docs/latest/gpu-collector/overview.mdx
index 1e8e6725..93052f59 100644
--- a/docs/latest/gpu-collector/overview.mdx
+++ b/docs/latest/gpu-collector/overview.mdx
@@ -1,6 +1,7 @@
---
-title: 'Overview'
-description: 'OpenTelemetry-native GPU and host metrics collector for NVIDIA, AMD, and Intel GPUs'
+title: 'GPU Monitoring for LLM Inference'
+sidebarTitle: "Overview"
+description: 'OpenTelemetry-native GPU monitoring for LLM inference on NVIDIA, AMD, and Intel GPUs — utilization, memory, power, and temperature.'
---
The **OpenTelemetry GPU Collector** is a lightweight, single-binary metrics collector written in Go. It exports GPU hardware telemetry, host system metrics, and process metrics via OpenTelemetry (OTLP) - with no Python dependencies, no DCGM daemon, and no vendor-specific agents.
diff --git a/docs/latest/openlit/chat/overview.mdx b/docs/latest/openlit/chat/overview.mdx
index 070395ed..535ca195 100644
--- a/docs/latest/openlit/chat/overview.mdx
+++ b/docs/latest/openlit/chat/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Otter is OpenLIT's AI chat assistant for observability - ask natural language questions about traces, costs, and tokens, and manage resources through conversation"
+title: "Otter AI Copilot for Observability"
+sidebarTitle: "Overview"
+description: "Otter is OpenLIT's AI chat assistant for observability — ask natural language questions about traces, costs, and tokens."
---
**Otter** is OpenLIT's AI chat assistant for observability data. Ask questions in plain language about traces, costs, and tokens; create dashboards, rules, prompts, and Vault secrets through conversation; and run the same AI analysis you get from the telemetry detail views. Product copy describes it as: *Ask questions about your observability data using natural language*.
diff --git a/docs/latest/openlit/coding-agents/overview.mdx b/docs/latest/openlit/coding-agents/overview.mdx
index 36eab3c0..a494c581 100644
--- a/docs/latest/openlit/coding-agents/overview.mdx
+++ b/docs/latest/openlit/coding-agents/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "OpenLIT ships first-class observability for AI coding agents - Claude Code, Cursor, Codex, and Windsurf - with no SDK and no code changes"
+title: "Coding Agent Observability: Claude Code, Codex, Cursor"
+sidebarTitle: "Overview"
+description: 'Coding agent observability for Claude Code, Cursor, Codex, and Windsurf — tokens, cost, tool calls, and outcomes with no SDK.'
---
OpenLIT ships first-class observability for AI coding agents. The `openlit` CLI ingests events from each vendor's hook system, normalizes them onto OpenTelemetry's `gen_ai.*` and OpenLIT's `coding_agent.*` semantic conventions, and ships them to your OpenLIT OTLP endpoint. The [Agents page](/latest/openlit/observability/agents/overview)'s Coding Agents tab detects them automatically - no controller, no SDK, no code change in your repos.
diff --git a/docs/latest/openlit/costs/manage-models/overview.mdx b/docs/latest/openlit/costs/manage-models/overview.mdx
index d86d08ed..80b56ec7 100644
--- a/docs/latest/openlit/costs/manage-models/overview.mdx
+++ b/docs/latest/openlit/costs/manage-models/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Manage models on the Costs page - edit LLM providers and per-model token prices used for AI cost tracking across traces, Otter, OpenGround, and the SDK"
+title: "Manage LLM Model Pricing"
+sidebarTitle: "Overview"
+description: 'Manage LLM providers and per-model token prices used for AI cost tracking across traces, Otter, OpenGround, and the SDK.'
---
**Manage models** is the **Manage models** tab on **Costs** (`/costs?tab=models`) under Monitor. Every connected ClickHouse database is seeded with **16 providers** and a large built-in model catalog (90+ models). You can edit any price, add custom providers and models, import/export pricing JSON, and point the OpenLIT SDK at a public pricing URL so ingest-time cost tracking stays in sync with the UI.
diff --git a/docs/latest/openlit/costs/overview.mdx b/docs/latest/openlit/costs/overview.mdx
index 221eba8d..d5da81e5 100644
--- a/docs/latest/openlit/costs/overview.mdx
+++ b/docs/latest/openlit/costs/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "AI cost tracking and AI cost analytics in OpenLIT - monitor LLM spend, manage model pricing, and backfill missing costs from Monitor → Costs"
+title: "LLM Cost Tracking"
+sidebarTitle: "Overview"
+description: 'LLM cost tracking in OpenLIT: monitor spend per model, user, and agent run, manage model pricing, and backfill missing costs.'
---
**Costs** (`/costs`) is OpenLIT's AI cost tracking and analytics page under **Monitor**. It combines spend analytics, LLM model pricing (Manage models), and Auto Pricing configuration in one place - the old Pricing and Manage Models pages redirect here.
diff --git a/docs/latest/openlit/dashboards/overview.mdx b/docs/latest/openlit/dashboards/overview.mdx
index d3027c68..39ea17bd 100644
--- a/docs/latest/openlit/dashboards/overview.mdx
+++ b/docs/latest/openlit/dashboards/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Create powerful, interactive dashboards to monitor AI application performance, visualize telemetry data, and gain insights into your LLM operations with real-time analytics."
+title: "AI Observability Dashboards"
+sidebarTitle: "Overview"
+description: 'Create interactive dashboards to monitor AI application performance, visualize telemetry, and analyze LLM operations in real time.'
---
<iframe className="w-full aspect-video rounded-xl" src="https://www.youtube.com/embed/XOvhKAo9XG8?si=IDssTOxW-hmF62P2" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
diff --git a/docs/latest/openlit/dashboards/widgets/overview.mdx b/docs/latest/openlit/dashboards/widgets/overview.mdx
index 6bdca0dd..bcb99a17 100644
--- a/docs/latest/openlit/dashboards/widgets/overview.mdx
+++ b/docs/latest/openlit/dashboards/widgets/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Learn about OpenLIT's widget system for creating powerful data visualizations, from time series charts to statistical summaries and interactive tables."
+title: "Dashboard Widgets"
+sidebarTitle: "Overview"
+description: 'OpenLIT dashboard widgets for AI observability visualizations: time series charts, statistical summaries, and interactive tables.'
---
## Available widget types
diff --git a/docs/latest/openlit/developer-resources/vault/overview.mdx b/docs/latest/openlit/developer-resources/vault/overview.mdx
index f1b73d05..22362c30 100644
--- a/docs/latest/openlit/developer-resources/vault/overview.mdx
+++ b/docs/latest/openlit/developer-resources/vault/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Securely manage LLM API keys and other secrets in one place, then retrieve them at runtime from any application via SDK or API"
+title: "LLM API Key Management (Vault)"
+sidebarTitle: "Overview"
+description: 'Vault securely manages LLM API keys and secrets, then retrieves them at runtime from any application via SDK or API.'
---
Vault is OpenLIT's built-in secrets manager for LLM API keys and other sensitive values. Instead of pasting an API key into every codebase, `.env` file, or teammate's laptop, you store it once in Vault and let any application fetch it at runtime - decoupling key rotation from redeploys.
diff --git a/docs/latest/openlit/evaluations/overview.mdx b/docs/latest/openlit/evaluations/overview.mdx
index 643781bc..3cdc7305 100644
--- a/docs/latest/openlit/evaluations/overview.mdx
+++ b/docs/latest/openlit/evaluations/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "LLM evaluation in OpenLIT - AI evals for hallucination, bias, toxicity, and custom criteria under Monitor → Evaluations"
+title: "Agent Evaluations (LLM-as-a-Judge & Programmatic Evals)"
+sidebarTitle: "Overview"
+description: 'Agent evals in OpenLIT score production traces with LLM-as-a-judge, programmatic checks, and human feedback — then gate releases in CI.'
---
**Evaluations** (`/evaluations`) is OpenLIT's LLM evaluation surface under **Monitor**. Score production traces with an LLM-as-a-judge, enable built-in or custom evaluators, schedule Auto Evaluation, and review pass-rate analytics - or run the same criteria [programmatically via the SDK](/latest/sdk/quickstart-programmatic-evals) for offline CI/CD gates.
diff --git a/docs/latest/openlit/observability/agents/overview.mdx b/docs/latest/openlit/observability/agents/overview.mdx
index a0fcac07..b4c77bfe 100644
--- a/docs/latest/openlit/observability/agents/overview.mdx
+++ b/docs/latest/openlit/observability/agents/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "AI agent observability, built entirely from OpenTelemetry trace data - discover every agent and its live call graph with zero registration"
+title: "AI Agent Observability"
+sidebarTitle: "Overview"
+description: 'Open-source AI agent observability built from OpenTelemetry traces: discover every agent and its live call graph with zero registration.'
---
The **Agents** page (`/agents`) is OpenLIT's AI agent observability view: a single place to see every agent OpenLIT has observed, built entirely from your OpenTelemetry trace data - there's nothing to declare or register up front. As soon as an agent's spans (including tool calls, model calls, and system prompts) show up in your traces, OpenLIT surfaces it here. See [Setup](/latest/openlit/observability/agents/setup) for what it takes to get an agent showing up.
diff --git a/docs/latest/openlit/observability/telemetry/overview.mdx b/docs/latest/openlit/observability/telemetry/overview.mdx
index 7355e485..17c2455f 100644
--- a/docs/latest/openlit/observability/telemetry/overview.mdx
+++ b/docs/latest/openlit/observability/telemetry/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "LLM observability for traces, metrics, logs, and exceptions - all in one Telemetry page, built on OpenTelemetry"
+title: "LLM Observability: Traces, Metrics & Logs"
+sidebarTitle: "Overview"
+description: 'LLM observability for traces, metrics, logs, and exceptions in one Telemetry page, built on OpenTelemetry.'
---
<video
diff --git a/docs/latest/openlit/prompts-experiments/context/overview.mdx b/docs/latest/openlit/prompts-experiments/context/overview.mdx
index 931b180a..5a90e153 100644
--- a/docs/latest/openlit/prompts-experiments/context/overview.mdx
+++ b/docs/latest/openlit/prompts-experiments/context/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "RAG context management for OpenLIT - store reusable content once and have the Rule Engine return the right piece automatically"
+title: "Context Management for RAG Pipelines"
+sidebarTitle: "Overview"
+description: 'RAG context management in OpenLIT: store reusable content once and have the Rule Engine return the right piece automatically.'
---
Context is OpenLIT's answer to RAG context management: a place to store reusable content for your RAG pipelines and AI applications, and have it returned automatically at runtime instead of hardcoding lookup logic in your application code.
diff --git a/docs/latest/openlit/prompts-experiments/openground/overview.mdx b/docs/latest/openlit/prompts-experiments/openground/overview.mdx
index 6a0220b2..7d55543c 100644
--- a/docs/latest/openlit/prompts-experiments/openground/overview.mdx
+++ b/docs/latest/openlit/prompts-experiments/openground/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "OpenGround is OpenLIT's AI model comparison tool - an LLM playground for running one prompt across many providers at once and comparing cost, speed, and quality"
+title: "Compare LLMs Side by Side (OpenGround)"
+sidebarTitle: "Overview"
+description: 'OpenGround compares LLMs side by side: run one prompt across many providers and compare cost, speed, and quality.'
---
OpenGround is OpenLIT's **AI model comparison tool** - an LLM playground built for a specific job: run **one prompt across many providers and models at once**, then compare the results side-by-side. Pick a prompt, pick the providers you want to test, click **Evaluate providers**, and see cost, speed, and quality laid out next to each other. Every run is saved automatically to a searchable history, so you can build up a record of what you've tested and reopen or reload any past run later.
diff --git a/docs/latest/openlit/prompts-experiments/prompt-hub/overview.mdx b/docs/latest/openlit/prompts-experiments/prompt-hub/overview.mdx
index 81d7a59e..f9eeb39d 100644
--- a/docs/latest/openlit/prompts-experiments/prompt-hub/overview.mdx
+++ b/docs/latest/openlit/prompts-experiments/prompt-hub/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Prompt Hub is OpenLIT's prompt management platform for versioning, publishing, and fetching prompts with dynamic variables at runtime"
+title: "Prompt Management with Prompt Hub"
+sidebarTitle: "Overview"
+description: 'Open-source prompt management with Prompt Hub: version, publish, and fetch prompts at runtime, linked to traces and evals.'
---
Prompt Hub is OpenLIT's prompt management platform: a place to create prompts, save them as explicit versions, and fetch them at runtime from your application instead of hardcoding prompt text in your codebase. It's built for teams who want prompt changes to ship independently of a code deploy, with a full history of what changed and when.
diff --git a/docs/latest/openlit/prompts-experiments/rule-engine/overview.mdx b/docs/latest/openlit/prompts-experiments/rule-engine/overview.mdx
index 46f24125..e3b1b0ec 100644
--- a/docs/latest/openlit/prompts-experiments/rule-engine/overview.mdx
+++ b/docs/latest/openlit/prompts-experiments/rule-engine/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Rule Engine is a rules engine for conditional AI resource retrieval - define matching conditions once, and let your application or OpenLIT ask what applies right now"
+title: "Rule Engine for Conditional Prompt & Context Retrieval"
+sidebarTitle: "Overview"
+description: 'Rule Engine is a rules engine for conditional AI resource retrieval — define matching conditions once and ask what applies at runtime.'
---
Rule Engine lets you define matching conditions once, instead of hardcoding lookup logic like "if model is X, use prompt Y" throughout your application. A rule pairs a set of conditions on runtime fields - like the model being called, the deployment environment, or the cost of a request - with the resources that should come back when those conditions are met. Your application calls a single API to ask "what applies right now?", and OpenLIT itself uses the same mechanism when retrieving a [Context](/latest/openlit/prompts-experiments/context/overview) or [Prompt](/latest/openlit/prompts-experiments/prompt-hub/overview) that's linked to a matching rule.
diff --git a/docs/latest/overview.mdx b/docs/latest/overview.mdx
index 696c873b..97f52a70 100644
--- a/docs/latest/overview.mdx
+++ b/docs/latest/overview.mdx
@@ -1,10 +1,26 @@
---
-title: "OpenLIT Overview"
+title: "What is OpenLIT? Open-Source Agent Harness Engineering Platform"
sidebarTitle: Overview
-description: "OpenLIT is an open-source AI Engineering platform for LLM observability, evaluations, prompt management, and cost tracking - built on OpenTelemetry, fully self-hostable."
+description: "OpenLIT is the open-source agent harness engineering platform: agent observability, evals, guardrails and prompt management on OpenTelemetry. Self-host free."
---
-Understand what your AI agents are doing, measure their performance, and improve them in production.
-With OpenLIT teams can Trace LLM and agent calls, compare models, iterate on prompts, detect regressions, and optimize cost and performance using real production data.
+
+**OpenLIT is an open-source agent harness engineering platform.** It gives teams OpenTelemetry-native tracing, evaluations, guardrails, prompt and context management, and cost and GPU monitoring for the harness around their AI agents and coding agents, so every agent failure can be traced, scored, and turned into a harness fix. OpenLIT is free to self-host under Apache 2.0 and works with any model, framework, or harness (Claude Code, Codex, OpenAI Agents SDK, LangGraph, CrewAI, and more).
+
+## What is agent harness engineering?
+
+An AI agent is a **model plus a harness**. The harness is everything except the model: tools, context, prompts, memory, hooks, guardrails, and feedback loops. **Agent harness engineering** is the discipline of designing, measuring, and improving that harness so agents are reliable in production.
+
+| Harness engineering step | OpenLIT feature |
+|---|---|
+| **Observe** every LLM call, tool call, MCP request, retrieval and agent step | [Agent observability](/latest/openlit/observability/agents/overview) & OpenTelemetry LLM tracing |
+| **Evaluate** quality, safety and cost on real traces | [Agent evals](/latest/openlit/evaluations/overview) (LLM-as-a-judge, programmatic, human feedback) |
+| **Guard** the agent at runtime | [Guardrails](/latest/sdk/features/guardrails) (prompt injection, sensitive topics, topic restriction) |
+| **Fix** the harness without redeploying | [Prompt Hub](/latest/openlit/prompts-experiments/prompt-hub/overview), [Context](/latest/openlit/prompts-experiments/context/overview), [Rule Engine](/latest/openlit/prompts-experiments/rule-engine/overview), [Vault](/latest/openlit/developer-resources/vault/overview) |
+| **Compare** models and prompts before shipping | [OpenGround](/latest/openlit/prompts-experiments/openground/overview) |
+| **Account** for cost, latency and hardware | [Cost tracking](/latest/openlit/costs/overview), [GPU monitoring](/latest/gpu-collector/overview) |
+
+See the [Concepts glossary](/latest/concepts/glossary) for short definitions of agent harness, agent observability, agent evals, guardrails, and related terms.
+
<Tabs>
<Tab title="Observe" icon="bars-staggered">
<video
@@ -139,16 +155,21 @@ With OpenLIT teams can Trace LLM and agent calls, compare models, iterate on pro
<AccordionGroup>
<Accordion title="What is OpenLIT?">
- OpenLIT is an open-source AI Engineering platform. It traces LLM and agent calls via OpenTelemetry, tracks cost and token usage, runs automated evaluations, and manages prompts - all self-hostable.
+ OpenLIT is an open-source agent harness engineering platform. It provides OpenTelemetry-native tracing, evaluations, guardrails, prompt management, and cost and GPU monitoring for AI agents and coding agents, and it is free to self-host under Apache 2.0.
+ </Accordion>
+ <Accordion title="What is an agent harness?">
+ An agent harness is everything in an AI agent except the model: the tools, context, prompts, memory, hooks, guardrails, and feedback loops that turn a model into a working agent. Claude Code, Codex, and frameworks such as LangGraph or CrewAI are harnesses; OpenLIT observes, evaluates, and improves any of them via OpenTelemetry.
+ </Accordion>
+ <Accordion title="What is agent harness engineering?">
+ Agent harness engineering is the discipline of designing, measuring, and improving everything around the model in an AI agent so the agent is reliable in production. Teams observe failures in traces, evaluate them, fix the harness, and verify the fix. OpenLIT is the open-source platform for that loop.
</Accordion>
<Accordion title="Is OpenLIT open source?">
- Yes. OpenLIT is fully open source and self-hostable via Docker Compose or Helm, so your telemetry and prompts never have to leave your infrastructure.
+ Yes. OpenLIT is fully open source under Apache 2.0 and self-hostable via Docker Compose or Helm, so your telemetry and prompts never have to leave your infrastructure.
</Accordion>
<Accordion title="Does OpenLIT require code changes to instrument my app?">
- No. OpenLIT's SDK auto-instruments 90+ LLMs, agent frameworks, and vector databases with zero code changes, or you can call `openlit.init()` once for manual instrumentation if you want more control - both produce the same OpenTelemetry traces.
+ No. OpenLIT's SDK auto-instruments 70+ LLMs, agent frameworks, and vector databases with zero code changes, or you can call `openlit.init()` once for manual instrumentation if you want more control - both produce the same OpenTelemetry traces.
</Accordion>
- <Accordion title="How is OpenLIT different from other LLM observability tools?">
- OpenLIT combines AI observability, evaluation, cost tracking, and prompt management in one open-source platform, instead of requiring separate tools for tracing, evals, and prompt versioning.
+ <Accordion title="Can OpenLIT send data to my existing observability stack?">
+ Yes. Because OpenLIT is OpenTelemetry-native, you can export traces and metrics to Grafana, Datadog, New Relic, SigNoz, or any OTLP-compatible backend - see [Destinations](/latest/sdk/destinations/overview).
</Accordion>
</AccordionGroup>
-
diff --git a/docs/latest/sdk/destinations/overview.mdx b/docs/latest/sdk/destinations/overview.mdx
index 19843591..2e5fb02d 100644
--- a/docs/latest/sdk/destinations/overview.mdx
+++ b/docs/latest/sdk/destinations/overview.mdx
@@ -1,6 +1,7 @@
---
-title: "Overview"
-description: "Send AI observability data to your existing observability stack"
+title: "Export OpenTelemetry LLM Traces to Any Backend"
+sidebarTitle: "Overview"
+description: 'Send OpenLIT OpenTelemetry traces and metrics to Grafana, Datadog, New Relic, SigNoz, or any OTLP backend.'
---
The OpenLIT SDK can send AI observability data (traces, metrics, logs) from LLMs, vector databases, and AI frameworks directly to your existing observability stack. This enables:
diff --git a/docs/latest/sdk/features/guardrails.mdx b/docs/latest/sdk/features/guardrails.mdx
index b6212be4..ba8d0b32 100644
--- a/docs/latest/sdk/features/guardrails.mdx
+++ b/docs/latest/sdk/features/guardrails.mdx
@@ -1,6 +1,6 @@
---
-title: 'Guardrails'
-description: 'Secure your app from Prompt Injection, Sensitive Topics, and Topic Restriction'
+title: 'LLM Guardrails: Prompt Injection, Sensitive Topics'
+description: 'Open-source LLM guardrails for prompt injection, sensitive topics, and topic restriction — traced with OpenTelemetry.'
---
## Overview
diff --git a/docs/latest/sdk/overview.mdx b/docs/latest/sdk/overview.mdx
index caa94c7e..96a361c7 100644
--- a/docs/latest/sdk/overview.mdx
+++ b/docs/latest/sdk/overview.mdx
@@ -1,6 +1,7 @@
---
-title: 'Overview'
-description: 'Production-ready AI observability with zero code changes. Monitor LLM applications, track token usage, detect hallucinations, and optimize costs with OpenTelemetry-native instrumentation.'
+title: 'OpenLIT SDK: OpenTelemetry Auto-Instrumentation for LLMs & Agents'
+sidebarTitle: "Overview"
+description: 'OpenTelemetry-native OpenLIT SDK for auto-instrumenting LLMs, agents, vector DBs, and GPUs — tracing, evals, and guardrails with one line of code.'
---
The OpenLIT SDK is an **open-source**, **OpenTelemetry-native** AI observability platform
diff --git a/opentelemetry-gpu-collector/README.md b/opentelemetry-gpu-collector/README.md
index e3b5e5d3..02c0655b 100644
--- a/opentelemetry-gpu-collector/README.md
+++ b/opentelemetry-gpu-collector/README.md
@@ -2,7 +2,7 @@
<img src="https://github.com/openlit/.github/blob/main/profile/assets/wide-logo-no-bg.png?raw=true" alt="OpenLIT Logo" width="30%"><h1>
OpenTelemetry GPU Collector</h1>
-**[Documentation](https://docs.openlit.io/latest/features/gpu) | [Quickstart](#-getting-started) | [Metrics](#-metrics) | [Configuration](#-configuration)**
+**[Documentation](https://docs.openlit.io/latest/gpu-collector/overview) | [Quickstart](#-getting-started) | [Metrics](#-metrics) | [Configuration](#-configuration)**
**[Roadmap](#%EF%B8%8F-roadmap) | [Feature Request](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Araised_hand%3A+Up+for+Grabs%2C+%3Arocket%3A+Feature&projects=&template=feature-request.md&title=%5BFeat%5D%3A) | [Report a Bug](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Abug%3A+Bug%2C+%3Araised_hand%3A+Up+for+Grabs&projects=&template=bug.md&title=%5BBug%5D%3A)**
@@ -16,7 +16,7 @@ OpenTelemetry GPU Collector</h1>
</div>
-A high-performance host and GPU metrics collector written in Go. Exports host-level system metrics (CPU, memory, disk, network), process metrics, DCGM-style GPU hardware telemetry, and eBPF-based CUDA kernel tracing (on by default on Linux) — all via OpenTelemetry (OTLP).
+Part of OpenLIT, the open-source agent harness engineering platform. A high-performance host and GPU metrics collector written in Go. Exports host-level system metrics (CPU, memory, disk, network), process metrics, DCGM-style GPU hardware telemetry, and eBPF-based CUDA kernel tracing (on by default on Linux) — all via OpenTelemetry (OTLP).
Metric names and attributes follow the [OpenTelemetry semantic conventions for hardware](https://opentelemetry.io/docs/specs/semconv/hardware/gpu/) and [system metrics](https://opentelemetry.io/docs/specs/semconv/system/).
diff --git a/sdk/go/README.md b/sdk/go/README.md
index 508539c1..e658b1e5 100644
--- a/sdk/go/README.md
+++ b/sdk/go/README.md
@@ -1,6 +1,6 @@
# OpenLIT Go SDK
-OpenTelemetry-native observability SDK for LLM applications in Go. Monitor your AI applications with automatic instrumentation for popular LLM providers.
+OpenTelemetry-native SDK for OpenLIT, the open-source agent harness engineering platform. Instrument LLM applications in Go to trace, evaluate, and improve the harness around your models — with automatic instrumentation for popular LLM providers.
[![Go Reference](https://pkg.go.dev/badge/github.com/openlit/openlit/sdk/go.svg)](https://pkg.go.dev/github.com/openlit/openlit/sdk/go)
[![Go Report Card](https://goreportcard.com/badge/github.com/openlit/openlit/sdk/go)](https://goreportcard.com/report/github.com/openlit/openlit/sdk/go)
diff --git a/sdk/python/README.md b/sdk/python/README.md
index 82bdcb7d..cdbffc7e 100644
--- a/sdk/python/README.md
+++ b/sdk/python/README.md
@@ -1,7 +1,7 @@
<div align="center">
<img src="https://github.com/openlit/.github/blob/main/profile/assets/wide-logo-no-bg.png?raw=true" alt="OpenLIT Logo" width="30%">
-<h3>OpenTelemetry-native</h3>
-<h1>AI Observability, Evaluation and Guardrails Framework</h1>
+<h3>OpenTelemetry-native SDK for</h3>
+<h1>OpenLIT — the open-source agent harness engineering platform</h1>
**[Documentation](https://docs.openlit.io/) | [Quickstart](#-getting-started-with-llm-observability) | [Roadmap](#️-roadmap) | [Feature Request](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Araised_hand%3A+Up+for+Grabs%2C+%3Arocket%3A+Feature&projects=&template=feature-request.md&title=%5BFeat%5D%3A) | [Report a Bug](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Abug%3A+Bug%2C+%3Araised_hand%3A+Up+for+Grabs&projects=&template=bug.md&title=%5BBug%5D%3A)**
@@ -19,9 +19,9 @@
</div>
-OpenLIT SDK is a monitoring framework built on top of **OpenTelemetry** that gives your complete Observability for your AI stack, from LLMs to vector databases and GPUs, with just one line of code with tracing and metrics. It also allows you to send the generated traces and metrics to your existing monitoring tools like Grafana, New Relic, and more.
+OpenLIT is the open-source agent harness engineering platform. This Python SDK is the OpenTelemetry-native auto-instrumentation library for tracing, evaluating, and guarding LLMs, agents, vector databases, and GPUs — with one line of code. Send traces and metrics to OpenLIT or to existing tools like Grafana, Datadog, and New Relic.
-This project proudly follows and maintains the [Semantic Conventions](https://github.com/open-telemetry/semantic-conventions/tree/main/docs/gen-ai) with the OpenTelemetry community, consistently updating to align with the latest standards in Observability.
+The SDK follows and maintains the [OpenTelemetry GenAI Semantic Conventions](https://github.com/open-telemetry/semantic-conventions/tree/main/docs/gen-ai) with the OpenTelemetry community.
## ⚡ Features
@@ -34,52 +34,52 @@ This project proudly follows and maintains the [Semantic Conventions](https://gi
| LLMs | Vector DBs | Frameworks | GPUs |
|--------------------------------------------------------------------------|----------------------------------------------|-------------------------------------------------|---------------|
-| [✅ OpenAI](https://docs.openlit.io/latest/integrations/openai) | [✅ ChromaDB](https://docs.openlit.io/latest/integrations/chromadb) | [✅ Langchain](https://docs.openlit.io/latest/integrations/langchain) | [✅ NVIDIA](https://docs.openlit.io/latest/integrations/nvidia-gpu) |
-| [✅ Ollama](https://docs.openlit.io/latest/integrations/ollama) | [✅ Pinecone](https://docs.openlit.io/latest/integrations/pinecone) | [✅ LiteLLM](https://docs.openlit.io/latest/integrations/litellm) | [✅ AMD](https://docs.openlit.io/latest/integrations/amd-gpu) |
-| [✅ Anthropic](https://docs.openlit.io/latest/integrations/anthropic) | [✅ Qdrant](https://docs.openlit.io/latest/integrations/qdrant) | [✅ LlamaIndex](https://docs.openlit.io/latest/integrations/llama-index) | |
-| [✅ GPT4All](https://docs.openlit.io/latest/integrations/gpt4all) | [✅ Milvus](https://docs.openlit.io/latest/integrations/milvus) | [✅ Haystack](https://docs.openlit.io/latest/integrations/haystack) | |
-| [✅ Cohere](https://docs.openlit.io/latest/integrations/cohere) | [✅ AstraDB](https://docs.openlit.io/latest/integrations/astradb) | [✅ EmbedChain](https://docs.openlit.io/latest/integrations/embedchain) | |
-| [✅ Mistral](https://docs.openlit.io/latest/integrations/mistral) | | [✅ Guardrails](https://docs.openlit.io/latest/integrations/guardrails) | |
-| [✅ Azure OpenAI](https://docs.openlit.io/latest/integrations/azure-openai) | | [✅ CrewAI](https://docs.openlit.io/latest/integrations/crewai) | |
-| [✅ Azure AI Inference](https://docs.openlit.io/latest/integrations/azure-ai-inference) | | [✅ DSPy](https://docs.openlit.io/latest/integrations/dspy) | |
-| [✅ GitHub AI Models](https://docs.openlit.io/latest/integrations/github-models) | | [✅ AG2](https://docs.openlit.io/latest/integrations/ag2) | |
-| [✅ HuggingFace Transformers](https://docs.openlit.io/latest/integrations/huggingface) | | [✅ Dynamiq](https://docs.openlit.io/latest/integrations/dynamiq) | |
-| [✅ Amazon Bedrock](https://docs.openlit.io/latest/integrations/bedrock) | | [✅ Phidata](https://docs.openlit.io/latest/integrations/phidata) | |
-| [✅ AI21](https://docs.openlit.io/latest/integrations/ai21) | | [✅ mem0](https://docs.openlit.io/latest/integrations/mem0) | |
-| [✅ Vertex AI](https://docs.openlit.io/latest/integrations/vertexai) | | [✅ MultiOn](https://docs.openlit.io/latest/integrations/multion) | |
-| [✅ Groq](https://docs.openlit.io/latest/integrations/groq) | | [✅ Julep AI](https://docs.openlit.io/latest/integrations/julep-ai) | |
-| [✅ ElevenLabs](https://docs.openlit.io/latest/integrations/elevenlabs) | | [✅ ControlFlow](https://docs.openlit.io/latest/integrations/controlflow) | |
-| [✅ vLLM](https://docs.openlit.io/latest/integrations/vllm) | | [✅ Crawl4AI](https://docs.openlit.io/latest/integrations/crawl4ai) | |
-| [✅ OLA Krutrim](https://docs.openlit.io/latest/integrations/krutrim) | | [✅ FireCrawl](https://docs.openlit.io/latest/integrations/firecrawl) | |
-| [✅ Google AI Studio](https://docs.openlit.io/latest/integrations/google-ai-studio) | | [✅ Letta](https://docs.openlit.io/latest/integrations/letta) | |
-| [✅ NVIDIA NIM](https://docs.openlit.io/latest/integrations/nvidia-nim) | | [✅ SwarmZero](https://docs.openlit.io/latest/integrations/swarmzero) | |
-| [✅ Titan ML](https://docs.openlit.io/latest/integrations/titan-ml) | | | |
-| [✅ Reka AI](https://docs.openlit.io/latest/integrations/reka) | | | |
-| [✅ xAI](https://docs.openlit.io/latest/integrations/xai) | | | |
-| [✅ Prem AI](https://docs.openlit.io/latest/integrations/premai) | | | |
-| [✅ Assembly AI](https://docs.openlit.io/latest/integrations/assemblyai) | | | |
-| [✅ Together](https://docs.openlit.io/latest/integrations/together) | | | |
-| [✅ DeepSeek](https://docs.openlit.io/latest/integrations/deepseek) | | | |
+| [✅ OpenAI](https://docs.openlit.io/latest/sdk/integrations/openai) | [✅ ChromaDB](https://docs.openlit.io/latest/sdk/integrations/chromadb) | [✅ Langchain](https://docs.openlit.io/latest/sdk/integrations/langchain) | [✅ NVIDIA](https://docs.openlit.io/latest/sdk/integrations/nvidia-gpu) |
+| [✅ Ollama](https://docs.openlit.io/latest/sdk/integrations/ollama) | [✅ Pinecone](https://docs.openlit.io/latest/sdk/integrations/pinecone) | [✅ LiteLLM](https://docs.openlit.io/latest/sdk/integrations/litellm) | [✅ AMD](https://docs.openlit.io/latest/sdk/integrations/amd-gpu) |
+| [✅ Anthropic](https://docs.openlit.io/latest/sdk/integrations/anthropic) | [✅ Qdrant](https://docs.openlit.io/latest/sdk/integrations/qdrant) | [✅ LlamaIndex](https://docs.openlit.io/latest/sdk/integrations/llama-index) | |
+| [✅ GPT4All](https://docs.openlit.io/latest/sdk/integrations/gpt4all) | [✅ Milvus](https://docs.openlit.io/latest/sdk/integrations/milvus) | [✅ Haystack](https://docs.openlit.io/latest/sdk/integrations/haystack) | |
+| [✅ Cohere](https://docs.openlit.io/latest/sdk/integrations/cohere) | [✅ AstraDB](https://docs.openlit.io/latest/sdk/integrations/astradb) | [✅ EmbedChain](https://docs.openlit.io/latest/sdk/integrations/overview) | |
+| [✅ Mistral](https://docs.openlit.io/latest/sdk/integrations/mistral) | | [✅ Guardrails](https://docs.openlit.io/latest/sdk/integrations/guardrails) | |
+| [✅ Azure OpenAI](https://docs.openlit.io/latest/sdk/integrations/azure-openai) | | [✅ CrewAI](https://docs.openlit.io/latest/sdk/integrations/crewai) | |
+| [✅ Azure AI Inference](https://docs.openlit.io/latest/sdk/integrations/azure-ai-inference) | | [✅ DSPy](https://docs.openlit.io/latest/sdk/integrations/dspy) | |
+| [✅ GitHub AI Models](https://docs.openlit.io/latest/sdk/integrations/github-models) | | [✅ AG2](https://docs.openlit.io/latest/sdk/integrations/ag2) | |
+| [✅ HuggingFace Transformers](https://docs.openlit.io/latest/sdk/integrations/huggingface) | | [✅ Dynamiq](https://docs.openlit.io/latest/sdk/integrations/dynamiq) | |
+| [✅ Amazon Bedrock](https://docs.openlit.io/latest/sdk/integrations/bedrock) | | [✅ Phidata](https://docs.openlit.io/latest/sdk/integrations/phidata) | |
+| [✅ AI21](https://docs.openlit.io/latest/sdk/integrations/ai21) | | [✅ mem0](https://docs.openlit.io/latest/sdk/integrations/mem0) | |
+| [✅ Vertex AI](https://docs.openlit.io/latest/sdk/integrations/vertexai) | | [✅ MultiOn](https://docs.openlit.io/latest/sdk/integrations/multion) | |
+| [✅ Groq](https://docs.openlit.io/latest/sdk/integrations/groq) | | [✅ Julep AI](https://docs.openlit.io/latest/sdk/integrations/julep-ai) | |
+| [✅ ElevenLabs](https://docs.openlit.io/latest/sdk/integrations/elevenlabs) | | [✅ ControlFlow](https://docs.openlit.io/latest/sdk/integrations/controlflow) | |
+| [✅ vLLM](https://docs.openlit.io/latest/sdk/integrations/vllm) | | [✅ Crawl4AI](https://docs.openlit.io/latest/sdk/integrations/crawl4ai) | |
+| [✅ OLA Krutrim](https://docs.openlit.io/latest/sdk/integrations/krutrim) | | [✅ FireCrawl](https://docs.openlit.io/latest/sdk/integrations/firecrawl) | |
+| [✅ Google AI Studio](https://docs.openlit.io/latest/sdk/integrations/google-ai-studio) | | [✅ Letta](https://docs.openlit.io/latest/sdk/integrations/letta) | |
+| [✅ NVIDIA NIM](https://docs.openlit.io/latest/sdk/integrations/nvidia-nim) | | [✅ SwarmZero](https://docs.openlit.io/latest/sdk/integrations/swarmzero) | |
+| [✅ Titan ML](https://docs.openlit.io/latest/sdk/integrations/titan-ml) | | | |
+| [✅ Reka AI](https://docs.openlit.io/latest/sdk/integrations/reka) | | | |
+| [✅ xAI](https://docs.openlit.io/latest/sdk/integrations/xai) | | | |
+| [✅ Prem AI](https://docs.openlit.io/latest/sdk/integrations/premai) | | | |
+| [✅ Assembly AI](https://docs.openlit.io/latest/sdk/integrations/assemblyai) | | | |
+| [✅ Together](https://docs.openlit.io/latest/sdk/integrations/together) | | | |
+| [✅ DeepSeek](https://docs.openlit.io/latest/sdk/integrations/deepseek) | | | |
| [✅ DigitalOcean (pydo)](https://docs.openlit.io/latest/sdk/integrations/digitalocean) | | | |
| [✅ DigitalOcean Gradient](https://docs.openlit.io/latest/sdk/integrations/digitalocean-gradient) | | | |
## Supported Destinations
-- [✅ OpenTelemetry Collector](https://docs.openlit.io/latest/connections/otelcol)
-- [✅ Prometheus + Tempo](https://docs.openlit.io/latest/connections/prometheus-tempo)
-- [✅ Prometheus + Jaeger](https://docs.openlit.io/latest/connections/prometheus-jaeger)
-- [✅ Grafana Cloud](https://docs.openlit.io/latest/connections/grafanacloud)
-- [✅ New Relic](https://docs.openlit.io/latest/connections/new-relic)
-- [✅ Elastic](https://docs.openlit.io/latest/connections/elastic)
-- [✅ Middleware.io](https://docs.openlit.io/latest/connections/middleware)
-- [✅ HyperDX](https://docs.openlit.io/latest/connections/hyperdx)
-- [✅ DataDog](https://docs.openlit.io/latest/connections/datadog)
-- [✅ SigNoz](https://docs.openlit.io/latest/connections/signoz)
-- [✅ OneUptime](https://docs.openlit.io/latest/connections/oneuptime)
-- [✅ Dynatrace](https://docs.openlit.io/latest/connections/dynatrace)
-- [✅ OpenObserve](https://docs.openlit.io/latest/connections/openobserve)
-- [✅ Highlight.io](https://docs.openlit.io/latest/connections/highlight)
-- [✅ SigLens](https://docs.openlit.io/latest/connections/siglens)
-- [✅ Oodle](https://docs.openlit.io/latest/connections/oodle)
+- [✅ OpenTelemetry Collector](https://docs.openlit.io/latest/sdk/destinations/otelcol)
+- [✅ Prometheus + Tempo](https://docs.openlit.io/latest/sdk/destinations/prometheus-tempo)
+- [✅ Prometheus + Jaeger](https://docs.openlit.io/latest/sdk/destinations/prometheus-jaeger)
+- [✅ Grafana Cloud](https://docs.openlit.io/latest/sdk/destinations/grafanacloud)
+- [✅ New Relic](https://docs.openlit.io/latest/sdk/destinations/new-relic)
+- [✅ Elastic](https://docs.openlit.io/latest/sdk/destinations/elastic)
+- [✅ Middleware.io](https://docs.openlit.io/latest/sdk/destinations/middleware)
+- [✅ HyperDX](https://docs.openlit.io/latest/sdk/destinations/hyperdx)
+- [✅ DataDog](https://docs.openlit.io/latest/sdk/destinations/datadog)
+- [✅ SigNoz](https://docs.openlit.io/latest/sdk/destinations/signoz)
+- [✅ OneUptime](https://docs.openlit.io/latest/sdk/destinations/oneuptime)
+- [✅ Dynatrace](https://docs.openlit.io/latest/sdk/destinations/dynatrace)
+- [✅ OpenObserve](https://docs.openlit.io/latest/sdk/destinations/openobserve)
+- [✅ Highlight.io](https://docs.openlit.io/latest/sdk/destinations/highlight)
+- [✅ SigLens](https://docs.openlit.io/latest/sdk/destinations/siglens)
+- [✅ Oodle](https://docs.openlit.io/latest/sdk/destinations/oodle)
## 💿 Installation
@@ -166,9 +166,9 @@ Configure the telemetry data destination as follows:
### Step 3: Visualize and Optimize!
Now that your LLM observability data is being collected and sent to configured OpenTelemetry destination, the next step is to visualize and analyze this data. This will help you understand your LLM application's performance and behavior and identify where it can be improved.
-If you want to use OpenLIT's Observability Dashboard to monitor LLM usage—like cost, tokens, and user interactions—please check out our [Quickstart Guide](https://docs.openlit.io/latest/quickstart).
+If you want to use OpenLIT's Observability Dashboard to monitor LLM usage—like cost, tokens, and user interactions—please check out our [Quickstart Guide](https://docs.openlit.io/latest/openlit/quickstart-ai-observability).
-If you're sending metrics and traces to other observability tools, take a look at our [Connections Guide](https://docs.openlit.io/latest/destinations/intro) to start using a pre-built dashboard we have created for these tools.
+If you're sending metrics and traces to other observability tools, take a look at our [Connections Guide](https://docs.openlit.io/latest/sdk/destinations/overview) to start using a pre-built dashboard we have created for these tools.
![](https://github.com/openlit/.github/blob/main/profile/assets/openlit-client-1.png?raw=true)
@@ -301,9 +301,9 @@ We are dedicated to continuously improving OpenLIT SDKs. Here's a look at what's
| [OpenTelmetry auto-instrumentation for LLM Providers like OpenAI, Anthropic]() | ✅ Completed |
| [OpenTelmetry auto-instrumentation for Vector databases like Pinecone, Chroma]() | ✅ Completed |
| [OpenTelmetry auto-instrumentation for LLM Frameworks like LangChain, LlamaIndex]() | ✅ Completed |
-| [OpenTelemetry-native auto-instrumentation for NVIDIA GPU Monitoring](https://docs.openlit.io/latest/features/gpu) | ✅ Completed |
-| [Real-Time Guardrails Implementation](https://docs.openlit.io/latest/features/guardrails) | ✅ Completed |
-| [Programmatic Evaluation for LLM Response](https://docs.openlit.io/latest/features/evaluations) | ✅ Completed |
+| [OpenTelemetry-native auto-instrumentation for NVIDIA GPU Monitoring](https://docs.openlit.io/latest/sdk/features/gpu) | ✅ Completed |
+| [Real-Time Guardrails Implementation](https://docs.openlit.io/latest/sdk/features/guardrails) | ✅ Completed |
+| [Programmatic Evaluation for LLM Response](https://docs.openlit.io/latest/sdk/features/evaluations) | ✅ Completed |
| [OpenTelemetry-native AI Agent Observability]() | ✅ Completed |
diff --git a/sdk/python/pyproject.toml b/sdk/python/pyproject.toml
index 8917a5dd..4b5f1c30 100644
--- a/sdk/python/pyproject.toml
+++ b/sdk/python/pyproject.toml
@@ -1,13 +1,13 @@
[tool.poetry]
name = "openlit"
version = "1.45.0"
-description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects"
+description = "OpenTelemetry-native SDK for OpenLIT, the open-source agent harness engineering platform: auto-instrument LLMs, agents, vector DBs and GPUs for tracing, evals and guardrails."
authors = ["OpenLIT"]
license = "Apache-2.0"
-repository = "https://github.com/openlit/openlit/tree/main/openlit/python"
+repository = "https://github.com/openlit/openlit"
readme = "README.md"
-homepage = "https://github.com/openlit/openlit/tree/main/openlit/python"
-keywords = ["OpenTelemetry", "otel", "otlp","llm", "tracing", "openai", "anthropic", "claude", "cohere", "llm monitoring", "observability", "monitoring", "gpt", "Generative AI", "chatGPT", "gpu"]
+homepage = "https://openlit.io"
+keywords = ["OpenTelemetry", "otel", "otlp", "llm", "tracing", "openai", "anthropic", "claude", "cohere", "llm monitoring", "observability", "monitoring", "gpt", "Generative AI", "chatGPT", "gpu", "agent-harness", "harness-engineering", "agent-observability", "llm-evaluation", "evals", "guardrails", "prompt-management", "ai-agents", "coding-agents", "mcp"]
packages = [
{ include = "openlit", from = "src" }
diff --git a/sdk/typescript/README.md b/sdk/typescript/README.md
index 5f7d6ebd..e5aebcaf 100644
--- a/sdk/typescript/README.md
+++ b/sdk/typescript/README.md
@@ -1,7 +1,7 @@
<div align="center">
<img src="https://github.com/openlit/.github/blob/main/profile/assets/wide-logo-no-bg.png?raw=true" alt="OpenLIT Logo" width="30%">
-<h3>OpenTelemetry-native</h3>
-<h1>AI Observability, Monitoring and Evaluation Framework</h1>
+<h3>OpenTelemetry-native SDK for</h3>
+<h1>OpenLIT — the open-source agent harness engineering platform</h1>
**[Documentation](https://docs.openlit.io/) | [Quickstart](#-getting-started-with-llm-observability) | [Roadmap](#️-roadmap) | [Feature Request](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Araised_hand%3A+Up+for+Grabs%2C+%3Arocket%3A+Feature&projects=&template=feature-request.md&title=%5BFeat%5D%3A) | [Report a Bug](https://github.com/openlit/openlit/issues/new?assignees=&labels=%3Abug%3A+Bug%2C+%3Araised_hand%3A+Up+for+Grabs&projects=&template=bug.md&title=%5BBug%5D%3A)**
@@ -18,9 +18,9 @@
</div>
-OpenLIT SDK is a monitoring framework built on top of **OpenTelemetry** that gives your complete Observability for your AI stack, from LLMs to vector databases, with just one line of code with tracing and metrics. It also allows you to send the generated traces and metrics to your existing monitoring tools like Grafana, New Relic, and more.
+OpenLIT is the open-source agent harness engineering platform. This TypeScript SDK is the OpenTelemetry-native auto-instrumentation library for tracing LLMs, agents, and vector databases — with one line of code. Send traces and metrics to OpenLIT or to existing tools like Grafana, Datadog, and New Relic.
-This project proudly follows and maintains the [Semantic Conventions](https://github.com/open-telemetry/semantic-conventions/tree/main/docs/gen-ai) with the OpenTelemetry community, consistently updating to align with the latest standards in Observability.
+The SDK follows and maintains the [OpenTelemetry GenAI Semantic Conventions](https://github.com/open-telemetry/semantic-conventions/tree/main/docs/gen-ai) with the OpenTelemetry community.
## ⚡ Features
diff --git a/sdk/typescript/package.json b/sdk/typescript/package.json
index 68d7c275..70b5360a 100644
--- a/sdk/typescript/package.json
+++ b/sdk/typescript/package.json
@@ -21,7 +21,7 @@
"LICENSE",
"package.json"
],
- "description": "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications, facilitating the integration of observability into your GenAI-driven projects",
+ "description": "OpenTelemetry-native SDK for OpenLIT, the open-source agent harness engineering platform: auto-instrument LLMs, agents, vector DBs and GPUs for tracing, evals and guardrails.",
"main": "dist/index.js",
"types": "dist/index.d.ts",
"exports": {
@@ -52,7 +52,17 @@
"monitoring",
"gpt",
"Generative AI",
- "chatGPT"
+ "chatGPT",
+ "agent-harness",
+ "harness-engineering",
+ "agent-observability",
+ "llm-evaluation",
+ "evals",
+ "guardrails",
+ "prompt-management",
+ "ai-agents",
+ "coding-agents",
+ "mcp"
],
"author": "Aman Agarwal <[email protected]>",
"license": "ISC",