Tip: select any text on this page to bring up the Ask tooltip and question the Synedrix OS assistant about it.
What LangGraph is
LangGraph describes itself as a low-level orchestration framework and runtime for building, managing, and deploying long-running, stateful agents. Its own documentation is emphatic about the level it operates at: it is very low-level and focused entirely on agent orchestration, and does not abstract prompts or architecture.
It provides the supporting infrastructure (durable execution, streaming, human-in-the-loop) for any long-running stateful workflow. A stated core strength is the ability to mix deterministic steps with LLM-driven agentic steps in a single graph.
The primitives are graphs of nodes and edges over a shared state object, with reducers controlling how updates merge, checkpointers persisting every super-step, and interrupts suspending a graph until external input arrives.
What it is normally used for
How LangGraph is used generally, before Synedrix OS enters the picture.
- Prompt chaining: a fixed sequence of model calls, optionally with a programmatic gate between steps
- Parallelization: fanning work out to several nodes in the same super-step and combining results
- Routing: classifying an input and directing it to a specialised downstream path
- Orchestrator-worker: an orchestrator node that plans and delegates to dynamically created workers
- Evaluator-optimizer: a generator paired with an evaluator that loops until the output passes
What it owns, and what it does not
Every platform in the stack has a non-overlapping responsibility. The boundary is as much a part of the design as the capability.
Owns
- Stateful agent graphs with checkpointing and durable execution
- The cyclic and multi-agent patterns: iterative, consensus, reflect-and-revise, hierarchical
- Human-in-the-loop interrupts that suspend and resume a graph at a checkpoint
- Tool-calling reasoning loops built with the LangChain library
Does not own
- Workflow triggers and business sequencing
- → n8n (Layer 3)
- Policy decisions
- → Policy Guard (os-agt-002)
- Tool authorization resolution
- → Orchestration passes in the active tool set
- LLM tracing, prompts, evaluations
- → Langfuse (Layer 10)
- Canonical business state
- → Supabase PG17 (Layer 7)
Why LangGraph
It is a runtime, not a workflow engine
An agent often needs to track what it's doing across many steps, and pick back up if something fails partway through. n8n runs steps in order and retries them; LangGraph does the harder job of holding that ongoing state and resuming from where it left off.
Pausing for human approval is built in
LangGraph can pause a run entirely and wait, for as long as needed, for a person to respond. That is exactly the shape of a human approval step, built into the runtime rather than bolted on afterwards.
Every step can be inspected and replayed
LangGraph saves a snapshot after every step, and any of those snapshots can be replayed or restarted from. A governed system needs to be able to show exactly what the agent knew at the moment it made a decision.
Agent behaviour is visible, not hidden in a prompt
Each agent's allowed behaviour is defined in the agent registryand built as a named, visible structure, not left implicit inside a prompt. That makes it possible to see an agent's boundaries just by looking at how it's built.
LangGraph questions, answered
Short, citable answers drawn from this page.
What role does LangGraph play in Synedrix OS?
LangGraph (Reasoning engine) is mapped to Layer 4: Agent Runtime in Synedrix OS. Where reasoning actually happens. It loops, accumulates state, and resumes.
What does LangGraph own in Synedrix OS?
LangGraph owns: Stateful agent graphs with checkpointing and durable execution; The cyclic and multi-agent patterns: iterative, consensus, reflect-and-revise, hierarchical; Human-in-the-loop interrupts that suspend and resume a graph at a checkpoint; Tool-calling reasoning loops built with the LangChain library.
What does LangGraph not own in Synedrix OS?
LangGraph does not own: Workflow triggers and business sequencing (owned by n8n (Layer 3)); Policy decisions (owned by Policy Guard (os-agt-002)); Tool authorization resolution (owned by Orchestration passes in the active tool set); LLM tracing, prompts, evaluations (owned by Langfuse (Layer 10)); Canonical business state (owned by Supabase PG17 (Layer 7)).
The full reference page, with the operating model, deployment detail, and links to every canonical specification this platform touches, lives in the documentation repository, which is private while the platform is being built.