langchain-ai/langgraph · error · ValueError

START cannot be an end node

Error message

START cannot be an end node

What it means

Raised by StateGraph.add_edge when the destination of an edge is the START sentinel. START is the graph's virtual entry point: it has no incoming edges, so it can never serve as a target. The guard at state.py:954 checks end_key == START after start_key has already been validated, meaning the start node was accepted but the end_key argument was mistakenly set to START (typically a swapped argument order or a copy-paste from an edge ending at END). The same check exists on the parallel branch for conditional-style edge lists at state.py:975.

Solutions

1. Target a real node name or END
2. Model loops by routing back to the first regular node, not START

When it happens

Trigger: Thrown at libs/langgraph/langgraph/graph/state.py:955 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of langchain-ai/langgraph@38031739e5 (2026-08-26). Data as JSON: /api/errors/5b7d192d0acb030b. Report an issue: GitHub.

Appendix: source

Thrown at libs/langgraph/langgraph/graph/state.py:955

            end_key: The key of the end node of the edge.

        Raises:
            ValueError: If the start key is `'END'` or if the start key or end key is not present in the graph.

        Returns:
            Self: The instance of the `StateGraph`, allowing for method chaining.
        """
        if self.compiled:
            logger.warning(
                "Adding an edge to a graph that has already been compiled. This will "
                "not be reflected in the compiled graph."
            )

        if isinstance(start_key, str):
            if start_key == END:
                raise ValueError("END cannot be a start node")
            if end_key == START:
                raise ValueError("START cannot be an end node")

            # run this validation only for non-StateGraph graphs
            if not hasattr(self, "channels") and start_key in set(
                start for start, _ in self.edges
            ):
                raise ValueError(
                    f"Already found path for node '{start_key}'.\n"
                    "For multiple edges, use StateGraph with an Annotated state key."
                )

            self.edges.add((start_key, end_key))
            return self

        for start in start_key:
            if start == END:
                raise ValueError("END cannot be a start node")
            if start not in self.nodes:
                raise ValueError(f"Need to add_node `{start}` first")

View on GitHub (pinned to 38031739e5)