Metadata-Version: 2.4
Name: agentpulse
Version: 0.1.0
Summary: AI Agent Observability SDK - Auto-instrument your AI agents with one line of code
Project-URL: Homepage, https://github.com/chkoutam/agentpulse
Project-URL: Documentation, https://github.com/chkoutam/agentpulse#readme
Project-URL: Repository, https://github.com/chkoutam/agentpulse
Project-URL: Issues, https://github.com/chkoutam/agentpulse/issues
Author-email: chkoutam <chkoutam@gmail.com>
License-Expression: Apache-2.0
Keywords: agents,ai,crewai,langgraph,llm,monitoring,observability,tracing
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development :: Libraries
Classifier: Topic :: System :: Monitoring
Requires-Python: >=3.10
Requires-Dist: httpx>=0.27
Requires-Dist: pydantic>=2.0
Requires-Dist: wrapt>=1.16
Provides-Extra: all
Requires-Dist: crewai>=0.80; extra == 'all'
Requires-Dist: langgraph>=0.2; extra == 'all'
Provides-Extra: crewai
Requires-Dist: crewai>=0.80; extra == 'crewai'
Provides-Extra: dev
Requires-Dist: pytest; extra == 'dev'
Requires-Dist: pytest-asyncio; extra == 'dev'
Provides-Extra: langgraph
Requires-Dist: langgraph>=0.2; extra == 'langgraph'
Description-Content-Type: text/markdown

# AgentPulse

Open-source observability SDK for AI agents. Auto-instrument CrewAI, LangGraph, and more with one line of code.

## Install

```bash
pip install agentpulse
```

With framework extras:

```bash
pip install agentpulse[crewai]
pip install agentpulse[langgraph]
pip install agentpulse[all]
```

## Quick Start

```python
import agentpulse
agentpulse.init()  # Auto-detects and instruments CrewAI, LangGraph

from crewai import Agent, Task, Crew
# Your existing code works unchanged — fully traced
```

## What it captures

- **Agent steps**: Which agent ran, with what goal
- **Task execution**: Task descriptions, expected vs actual output
- **Tool calls**: Tool name, input, output
- **LLM calls**: Model, tokens, cost, latency
- **Decisions**: Agent routing and choices

## Manual instrumentation

```python
from agentpulse import trace, tool

@trace(name="my-agent-step")
def process_data(data):
    ...

@tool(name="search")
def search_web(query: str):
    ...
```

## Dashboard

AgentPulse comes with a full web dashboard. See the [main repo](https://github.com/chkoutam/agentpulse) for setup instructions.

## License

Apache 2.0
