Skip to main content

Introduction

AgentSight is a conversation tracking and analytics platform built to provide your clients with access to their conversational AI data, including dashboards, transcripts, analytics overviews and more.

Unlike traditional observability platforms built for developers, AgentSight focuses on client visibility and meaningful insights, not just logs or traces.

Besides the client-facing platform, you also get a fully managed database backend for your conversational AI, so you do not need to build or maintain any infrastructure or dashboards, which lets you focus on your AI.

The Python SDK makes this integration possible. You will be sending your metrics and tracking in just a few lines of code.

Visit the landing page for more information.

What It's Used For

AgentSight’s core purpose is to help you share real-time conversation data, transcripts, and analytics directly with your clients.

Clients gain direct access to:

  • Conversation transcripts - see exactly how users interact
  • Usage analytics - engagement trends, token usage, performance metrics
  • Custom reports - build reports based on filters, metrics, timeframes
  • Data export - download or integrate the raw data for internal use

This transforms your offering from just building AI solutions to delivering a full, data-driven platform your clients can actively use.

Comparison with Observability Platforms

Feature AgentSight Langfuse Phoenix Arize
Primary Focus Conversation analytics & client dashboards AI Observability LLM tracing, evaluation
Designed For Clients & End Users Developers Developers
Client Dashboard Access ✅ ❌ ❌
Actionable performance reports ✅ ❌ ❌
Conversation analytics dashboard ✅ ❌ ❌
Shareable Transcripts ✅ ❌ ❌
White-Label Support ✅ ❌ ❌
Usage Metrics Tracking ✅ ❌ ❌
Usage Analytics & Reports ✅ ❌ ❌
Automatic Token & Cost Tracking ✅ ✅ ✅
Raw Trace Data via API ✅ ✅ ✅
Trace Debugging UI¹ ❌ ✅ ✅
Data Export ✅ ✅ ✅

¹ AgentSight records complete OpenTelemetry spans and makes them readable through its API, but its dashboards are built for client visibility, not internal debugging.

AgentSight complements observability platforms. It’s not built for debugging, but for giving clients insight into their own AI systems and providing developers a simple database and API for conversation persistence.

Installation

pip install agentsight

Or with your package manager of choice — poetry add agentsight, uv add agentsight.

Quick start

Track complete conversations in a few lines of code:

import agentsight

agentsight.init(api_key="ags_...")  # or set AGENTSIGHT_API_KEY

with agentsight.conversation("your_conversation_id"):
    with agentsight.turn():
        agentsight.user_message("What's the weather like today?")
        reply = my_agent.run(...)  # token usage, cost and tool calls captured automatically
        agentsight.agent_message(reply)

Sending happens in the background — there is nothing to flush in a long-running service. LLM calls made through OpenAI, Anthropic, LangChain or LlamaIndex inside a turn are captured without any extra code, and agentsight.upload_attachments(...) delivers files shared in the conversation.

Learn More

Visit the docs to learn more: docs

Metadata

Release files for agentsight 0.1.4

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for agentsight 0.1.4
File Size Uploaded Provenance Actions
agentsight-0.1.4.tar.gz 209.8 kB Without provenance Details for agentsight-0.1.4.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for agentsight 0.1.4
File Interpreter ABI Platform Actions
agentsight-0.1.4-py3-none-any.whl Python 3 none any Details for agentsight-0.1.4-py3-none-any.whl

Total release size: 367.0 kB

Release files / agentsight-0.1.4.tar.gz

Download URL agentsight-0.1.4.tar.gz
Size 209.8 kB
Tags Source
SHA-256 checksum
How to use checksums
f9c6e71dd3d67412b3db2d14db0a4deb89bcf5bfa004a7c8fbdbd5e6e03824f3
BLAKE2b-256 checksum
How to use checksums
7a725c1acee88f633759d3b243180acd8ae2525ba5054370852396c81b144331
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.11.0

Release files / agentsight-0.1.4-py3-none-any.whl

Download URL agentsight-0.1.4-py3-none-any.whl
Size 157.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
ae91de2b80c7234e866451b4c207acb1732ce0eb25d6c4ffd7121273bb45f25d
BLAKE2b-256 checksum
How to use checksums
106eb5f50d03aaa32b0d75b8ec3f204ac5286646721849f86843b209dec3477a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.11.0

Security for agentsight 0.1.4

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page