Recover Delta-Channel State from Postgres LangGraph Threads
Recovers delta-channel state from a Postgres-backed LangGraph thread, enabling state reconstruction and debugging of multi-step AI workflows.
Maintainer of this project? Claim this page to edit the listing.
1.0.0Add to Favorites
Why it matters
Extract and recover delta-channel state data from Postgres-backed LangGraph conversation threads for debugging, migration, or state inspection purposes.
Outcomes
What it gets done
Connect to Postgres database storing LangGraph thread data
Query and extract delta-channel state information from specific threads
Reconstruct conversation state from database records
Export recovered state data for analysis or restoration
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/lg-dump | bash Overview
Dump
This tool recovers delta-channel state from a Postgres-backed LangGraph thread. It extracts the complete state history of LangGraph workflows that use delta channels, enabling developers to reconstruct and inspect how state evolved throughout workflow execution. Use this when you need to debug LangGraph workflows stored in Postgres, recover from interrupted executions, or audit state transitions in completed workflows. It is essential when you need to understand what happened at each step of a workflow that failed or behaved unexpectedly.
What it does
This prompt chain recovers delta-channel state from a Postgres-backed LangGraph thread. It extracts and reconstructs the state history of LangGraph workflows that use delta channels for state management, allowing developers to inspect, debug, and recover workflow state from persistent storage.
When to use - and when NOT to
Use this when you need to debug a LangGraph workflow that has failed or behaved unexpectedly and you need to examine its state history. Use it when recovering from interrupted workflows where you need to understand what state was captured before the interruption. Use it when auditing or analyzing the state transitions of a completed workflow stored in Postgres.
Do not use this for live workflow monitoring - it is designed for post-hoc analysis of stored state. Do not use this if your LangGraph implementation does not use Postgres as a backend or does not employ delta channels for state management.
Inputs and outputs
You provide the connection details for your Postgres database and the thread identifier for the LangGraph workflow you want to inspect. The tool queries the Postgres backend where LangGraph persists its delta-channel state.
You receive the reconstructed state history from the delta channel, allowing you to see the progression of state changes throughout the workflow execution. This output enables you to understand what happened at each step of the workflow.
Integrations
Postgres: The tool connects directly to Postgres databases that serve as the persistence layer for LangGraph workflows, querying the tables where delta-channel state is stored.
LangGraph: Designed specifically for LangGraph workflows that use delta channels for state management, this tool understands the LangGraph state persistence format and can reconstruct the state timeline from stored deltas.
Who it's for
This tool is for developers building and maintaining LangGraph-based AI workflows who need to debug state-related issues. It serves DevOps engineers responsible for monitoring and recovering from workflow failures in production environments. Data engineers and AI engineers who need to audit or analyze the behavior of complex multi-step AI workflows will find this useful for understanding state transitions and identifying where workflows diverged from expected behavior.
Source code
Recover delta-channel state from a Postgres-backed LangGraph thread.
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.