Skip to content

Agent framework integrations (LangChain / AutoGen) #1

Description

@joshualamerton

AgentLens should integrate with common AI agent frameworks so developers can instrument their agents without modifying large parts of their code.

This issue focuses on creating lightweight wrappers for popular frameworks such as:

  • LangChain
  • AutoGen

The goal is to allow AgentLens to capture events such as:

• tool calls
• model prompts
• model responses
• agent state changes

Example usage concept:

from agentlens import instrument

agent = instrument(langchain_agent)

agent.run("Find the best laptop under $1500")

Expected outcome

Running the agent automatically emits AgentLens events such as:

tool_call
model_prompt
model_response
state_update

These events should appear in the AgentLens event stream.

Tasks

  • design framework wrapper interface
  • implement LangChain integration
  • implement AutoGen integration
  • add example demos in the examples folder

Success criteria

A developer should be able to instrument an existing agent with one line of code.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions