Agensight SDK
A simple SDK for tracing and evaluating AI agent workflows.
Installation
pip install --upgrade agensight
Quick Start
from agensight import init, trace, span
# Initialize Agensight
init(name="my-agent")
# Add tracing to your functions
@trace("my_workflow")
def my_function():
@span()
def my_subtask():
# Your code here
pass
return my_subtask()
Key Features
1. Tracing
- Auto-instrumented tracing of LLM calls
- Function execution tracking
- Agent interaction monitoring
- Performance metrics and token usage analytics
2. Evaluation
from agensight.eval.metrics import GEvalEvaluator
# Create custom evaluator
accuracy = GEvalEvaluator(
name="Accuracy",
criteria="Evaluate response accuracy",
threshold=0.7
)
# Use in your code
@span(metrics=[accuracy])
def my_function():
# Your code here
pass
3. Local Storage
- All data stored locally in
.agensight - No external data uploads
- Complete privacy and control
Dashboard
Start the dashboard to view traces and metrics:
agensight view
Visit http://localhost:5001 in your browser
Documentation
For detailed documentation, visit our docs.
Support
Join our GitHub Discussions for support and feature requests.
License
MIT License • © 2025 agensight contributors
2 maintainers
Metadata
2 maintainers
Release files for agensight 0.6.3
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 |
|---|---|---|---|---|
| agensight-0.6.3.tar.gz | 1.1 MB | Without provenance | Details for agensight-0.6.3.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agensight-0.6.3-py3-none-any.whl | Python 3 | none | any | Details for agensight-0.6.3-py3-none-any.whl |
Total release size: 2.3 MB
Release files / agensight-0.6.3.tar.gz
| Download URL | agensight-0.6.3.tar.gz |
|---|---|
| Size | 1.1 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
d52d9d0dddd2de711d7f51953439f7b6ad02844ff2670f62b71f85356401cac8
|
|
BLAKE2b-256 checksum How to use checksums |
ce0c196811b5d09806e6edb01b615aac37d27a08deb24678cafc7102e8975484
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Release files / agensight-0.6.3-py3-none-any.whl
| Download URL | agensight-0.6.3-py3-none-any.whl |
|---|---|
| Size | 1.2 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
61b34efd9d76a7c41806323bbf649762b155c4544bf67d067f37e0a6ebc8d1ff
|
|
BLAKE2b-256 checksum How to use checksums |
11942c3b88040cfeaaa517c2a1284aab1a7545970e11205af5a9ab8e6f434c8c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Security for agensight 0.6.3
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.
