🔭 AgentixLens
AI Agent Observability — trace, monitor, and debug your AI agents in real-time.
🌐 Website · 📊 Dashboard · 🐦 Twitter
What is AgentixLens?
AgentixLens gives you a transparent window into every decision your AI agents make.
- Visual Trace Explorer — every LLM call, tool use, and decision branch in a waterfall
- Cost & Token Tracking — per-run, per-model, per-tool granularity
- Latency Heatmaps — p50/p95/p99 across your entire agent fleet
- Failure Capture & Replay — reproduce any failed run with full context
- Model-Agnostic — works with any LLM, zero vendor lock-in
Repo Structure
agentixlens/
├── index.html # Landing page (deployed to Netlify)
├── dashboard.html # Dashboard UI (deployed to Netlify)
├── netlify.toml # Netlify config
├── backend/ # FastAPI backend (deploy to Railway/Render/Fly)
│ ├── main.py
│ ├── requirements.txt
│ ├── Dockerfile
│ ├── .env.example # Copy to .env and fill in values
│ ├── db/
│ ├── models/
│ ├── routers/
│ └── middleware/
└── sdk/ # Python SDK (publish to PyPI)
├── agentixlens/
├── examples/
└── tests/
Quick Start
1. Install the SDK
pip install agentixlens
2. Instrument your agent
from agentixlens import lens, trace
lens.init(project="my-agent")
@trace("research-agent")
async def run_agent(query: str) -> str:
# Your existing agent code — unchanged
...
3. Run the backend locally
cd backend
pip install -r requirements.txt
cp .env.example .env # Fill in your values
uvicorn main:app --reload --port 4317
4. Open the dashboard
Open dashboard.html in your browser — it connects to http://localhost:4317 automatically.
Deployment
Frontend (Netlify) — free
- Push this repo to GitHub
- Connect to Netlify → Import from Git
- Set Publish directory to
.(root) - Set Build command to (empty)
- Deploy —
index.htmlanddashboard.htmlgo live instantly
Backend (Railway) — one click
Or manually:
cd backend
railway login
railway up
Set these environment variables on Railway:
ENV=production
AUTH_ENABLED=true
API_SECRET_KEY=<generate with: python -c "import secrets; print(secrets.token_urlsafe(32))">
DASHBOARD_TOKEN=<generate same way>
ALLOWED_ORIGINS=https://agentixlens.com
Backend (Docker)
cd backend
docker build -t agentixlens-backend .
docker run -p 4317:4317 \
-e API_SECRET_KEY=your-key \
-e DASHBOARD_TOKEN=your-token \
-v agentixlens-data:/data \
agentixlens-backend
SDK Reference
from agentixlens import lens, trace, trace_llm, trace_tool, current_span
# Initialize
lens.init(
project="my-agent",
endpoint="https://your-backend.railway.app",
api_key="ax_...",
debug=True,
)
# Trace entire agent run
@trace("my-agent", tags={"env": "prod"})
async def run_agent(query: str): ...
# Trace individual LLM call
@trace_llm(model="claude-3-5-sonnet", provider="anthropic")
async def call_llm(messages): ...
# Trace tool calls
@trace_tool("web_search")
async def search(query: str): ...
# Add metadata from inside any function
current_span().set_attribute("user_id", user.id)
LangChain Integration
from agentixlens.integrations.langchain import AgentixLensCallback
llm = ChatOpenAI(callbacks=[AgentixLensCallback()])
Environment Variables
See backend/.env.example for all options.
| Variable | Required | Description |
|---|---|---|
API_SECRET_KEY |
Yes (prod) | SDK authentication key |
DASHBOARD_TOKEN |
Yes (prod) | Dashboard UI access token |
ENV |
No | development or production |
AGENTIXLENS_DB |
No | SQLite path (default: ~/.agentixlens/server.db) |
ALLOWED_ORIGINS |
No | Comma-separated CORS origins |
AUTH_ENABLED |
No | Set false for local dev |
Contributing
- Fork the repo
- Create a feature branch:
git checkout -b feature/my-feature - Run tests:
cd sdk && pytest tests/ -v - Commit and push
- Open a Pull Request
License
MIT © 2025 AgentixLens
Built for developers who are tired of flying blind with AI agents.
Metadata
Release files for agentixlens 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentixlens-0.1.0.tar.gz | 17.8 kB | Without provenance | Details for agentixlens-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentixlens-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for agentixlens-0.1.0-py3-none-any.whl |
Total release size: 35.5 kB
Release files / agentixlens-0.1.0.tar.gz
| Download URL | agentixlens-0.1.0.tar.gz |
|---|---|
| Size | 17.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
6536f71b4b805fb149eac916b9798dc974b6af12a3e76185ac54109d16da8748
|
|
BLAKE2b-256 checksum How to use checksums |
aa9568a905d86b4e73a6e229036bc669baf9937fb6233e7af671cc1c61b0b33a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.8
|
Release files / agentixlens-0.1.0-py3-none-any.whl
| Download URL | agentixlens-0.1.0-py3-none-any.whl |
|---|---|
| Size | 17.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c04907f0954a49bcd6150b4105cc9dd8babbcb993189b2e2fca7aa8e7c23c0a8
|
|
BLAKE2b-256 checksum How to use checksums |
7d229a41b8e6631c9d260379eb9f8a0180a9c38895abfe497bc3abeb37eb7ffc
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.8
|
Security for agentixlens 0.1.0
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.