agentops
agentops copied to clipboard
Python SDK for agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks like CrewAI, Langchain, and Autogen
## 🚀 Feature Request **💡 Got a brilliant idea?** Support the official Ollama python library by instrumenting call code. Update `llm_tracker.py` to search for ollama
## 🐛 Bug Report Several of the Jupyter notebooks don't run end-to-end with the latest version of AgentOps. Additionally, there are a handful of spelling/grammar issues. To improve this: 1....
## 📥 Pull Request **📘 Description** 1. Added self.record(event) before we try to get the return value of a function that is decorated 2. Modified the worker add_event function such...
Currently we don't see the last termination event. We should emit that. So for example, ctrl+C should show that termination event
General nit, why are the docstrings mis-aligned with the triple quotes? I'm seeing: ``` def foo(): """ stuff stuff stuff """ ``` vs. ``` def foo(): """ stuff stuff stuff...
## 🚀 Feature Request **💡 Got a brilliant idea?** Currently, our test coverage is not being tracked. Also, we want to abide by Mypy typing **🤔 Is your feature request...
## 🚀 Feature Request **💡 Got a brilliant idea?** Using either OpenInference or OpenLLMetry, add an instrumented for DSPY. This is the first of several test features we'll use to...
## 🚀 Feature Request **💡 Got a brilliant idea?** Similar to Langchain, Llama Index also provides callback handlers. This is a potential opportunity to give built-in observability to agents built...
## 🐛 Bug Report **🔎 Describe the Bug** The time gap between events is inconsistent. I think we may have a blocking issue with how we post data. **🔄 Reproduction...
## 📥 Pull Request **📘 Description** This PR adds functions in the `llm_tracker.py` file for the asynchronous and synchronous Anthropic SDK support. An example notebook `anthropic_example` under `examples/anthropic` is also...