dkondo

agent-tackle-box: Debug LangGraph and LangChain Agents in the Terminal

agent-tackle-box is a Python toolkit centered on agent-debugger, a Textual terminal debugger for LangGraph and LangChain agents. It brings agent state and tool-call inspection together with Python breakpoints and stepping in one interface.

Stars
52
Forks
5
Stars, last 30 days
+1
Last commit
5 months ago
Contributors
2

Our take

Quietno commits in the last 90 days and no release in a year

Health score

21/100

How it is scored
Activity
19
Community
24
Issues
21
Pull requests
20

+1 stars in the last 30 days

  • Fewer contributors than 86% of the projects we track

This is a focused debugging toolkit for LangGraph and LangChain development, but its recent quiet period and concentrated contributor base warrant checking project activity before depending on it.

Good fit if

  • You build agents with LangGraph or LangChain and want agent and Python debugging in one terminal UI.
  • Inspecting state, tool calls, semantic breakpoints, or Python execution addresses a concrete debugging need for your team.
  • You are comfortable evaluating an MIT-licensed project with no recent commits or releases.

Look elsewhere if

  • Your agent stack is not based on LangGraph or LangChain.
  • You need an actively maintained project with recent commits or releases.
  • You require evidence of automated tests before adopting a debugging tool. The supplied project facts indicate CI but no tests.
All health signals
Last commit2026-05-11 (5 months ago)
Commits, last 90 days0
Releases, last 12 months0
Contributors2 (top contributor: 83% of commits)
Issues closed, last 90 days0
Pull requests merged, last 90 days0
Project age8 months

Checked on 2026-10-17 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

53

Forks

5

Change since Nov 17, 2026: 0 stars·0 forks

Overview

agent-tackle-box is a Python toolkit for developing AI agents, centered on agent-debugger, a Textual terminal debugger for LangGraph and LangChain agents. It addresses the gap between inspecting an agent's behavior and stepping through the Python code that drives it, bringing both into one interface.

The debugger is aimed at developers working with LangGraph or LangChain who need to examine agent state, messages, and tool calls, then investigate the corresponding execution in Python. The repository points to a dedicated project README for usage and extension details.

Key Features

  • Inspect agent state and messages in a terminal UI.
  • View tool calls and store snapshots.
  • Set semantic breakpoints for agent-level debugging.
  • Use Python line breakpoints and step through code.
  • Inspect the stack and local variables.
  • Combine agent-level and Python-level debugging in one Textual interface.

Use Cases

  • LangGraph developers tracing state changes and tool calls during an agent run.
  • LangChain developers investigating how Python execution relates to agent behavior.
  • Agent builders using semantic breakpoints to pause on meaningful agent events.
  • Developers diagnosing a run by inspecting its stack and local variables alongside agent state.

What you need

Detected in the repository

  • Python (main language on GitHub)
  • Automated checks on GitHub Actions

License in plain words

MITpermissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the copyright and license notice
  • Share your changes: no

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

See the agent-debugger README for installation, usage, commands, and extension details. The top-level README does not provide an installation command.

Considerations

The project is marked Quiet: it has had no commits in the last 90 days and no release in the past year. It has two contributors, with most contributions coming from one person, so maintenance depends substantially on a small contributor base. The supplied facts indicate CI but no tests; review the project and its dedicated README for your own adoption requirements.

Found this useful?

Share it with someone who would like agent-tackle-box.