langgraph-time-travel

Navigate LangGraph checkpoints to inspect history and replay execution.

3|1|Updated Jan 31, 2026
One-click install
npx skills add https://github.com/christian-bromann/langchain-skills --skill langgraph-time-travel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: langgraph-time-travel
Source: https://github.com/christian-bromann/langchain-skills/tree/main/skills/langgraph-time-travel/python
Command: npx skills add https://github.com/christian-bromann/langchain-skills --skill langgraph-time-travel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Time travel in LangGraph enables debugging and auditing agent decisions by letting you inspect past states, replay execution from checkpoints, fork execution paths, and diagnose issues with historical state.

Core Features & Use Cases

  • Inspect decision points and execution history using get_state_history
  • Replay execution from a checkpoint with optional new input
  • Fork threads/state at checkpoints to explore alternate paths
  • Update state before replay to test recovery or adjustments
  • Diagnose errors by inspecting checkpoints, metadata, and state transitions

Quick Start

Create a MemorySaver checkpointer, compile the graph, invoke with a thread_id, then inspect the state history to understand decision points.

Frequently Asked Questions about langgraph-time-travel

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I inspect past states and execution history in LangGraph?

Inspect past states in LangGraph by calling get_state_history on your compiled graph. This retrieves past checkpoints and metadata, allowing you to audit agent decisions and understand state transitions across threads.

Can I replay a LangGraph execution from a specific checkpoint?

Replay a LangGraph execution from a specific checkpoint by invoking graph.invoke with a configuration containing both thread_id and checkpoint_id. You can optionally provide new input to test different outcomes from that historical state.

How do I fork an alternative execution path from a checkpoint in LangGraph?

Fork alternative paths in LangGraph by retrieving a historical state with get_state, updating it with update_state, and resuming execution. This explores alternate scenarios without modifying the original execution thread.

What do I need to set up before using time travel debugging in LangGraph?

Time travel debugging requires a working checkpointer, such as MemorySaver, compiled into your graph. You also need access to get_state_history, get_state, and update_state APIs to navigate and manipulate historical states.

How can I update LangGraph state before resuming execution to test recovery?

Update LangGraph state before resuming by using the update_state method on a specific checkpoint. This allows you to adjust historical state variables and test recovery scenarios by replaying execution with the modified data.