[jeffdupont]

Summary

Fixes AIC-3592 (1.0 blocker) in Python. to_lang_graph failed every turn that handed off: the Whodunit app's check --engine langgraph failed 30 of 30 graph turns with ChatBedrockConverse._converse_params() got an unexpected keyword argument 'parallelToolCalls'. Implements the new §2.x.3 bullets in launchdarkly/ai-sdks-monorepo#118.

In packages/langchain-agents/src/launchdarkly_ai_langchain_agents/native_graph.py:

• parallel_tool_calls is no longer passed to bind_tools. This is the Bedrock error. The first-hand-off rule below replaces it.
• The hand-off Command answers its tool call. It now adds a ToolMessage (Handed off to <target>.) with the call's id. The id comes from InjectedToolCallId, so the hand-off schema is now a small pydantic model (a dict schema skips the injection). pydantic>=2.13 is declared, as in the other packages that import it.
• The routing node does not run again after a hand-off. The static tools -> node edge is replaced by a conditional edge that goes back to the node only when no hand-off routed. This applies to single-child and multi-child nodes.
• Only the first hand-off in a message routes, as graph() does. A later one returns Ignored: only the first transfer in a turn routes. and does not track $ld:ai:graph:handoff_success.
• A hand-off target sees provider-neutral text history. Messages already in the state when a node is entered are rewritten for its model: reasoning and other non-text blocks dropped, each assistant message's tool calls appended as [Tool call] <name> <args> lines, the tool results after it folded into one user message of [Tool result] <name>: <text> lines in call order. Human and system messages pass unchanged. The node's own tool loop after entry stays native.

A normal tool call in the same message as a hand-off still runs, in the same tools step, and the target receives its result (decision 1A on the ticket).

Tests

TestNativeHandoffRealStateGraph runs six cases on the real StateGraph with a mock model: a multi-child hand-off, a single-child hand-off, a normal call plus a hand-off, two hand-offs, no parallel_tool_calls, and a node's own tool loop staying native. Each hand-off case asserts the exact messages the target receives (no tool role, tool calls, or reasoning blocks). TestToNeutralHistory unit-tests the rewrite on one input covering every rule (reasoning, null text, an empty assistant, an unmatched result, call order, nested and non-ASCII args, a message chunk, an image in a human message); the TypeScript test uses the same input and expects the same output.

Local: make lint, make format-check, make typecheck, make test (3718 passed, 18 skipped) all pass. The live Bedrock run is under "Live check".

TypeScript: launchdarkly/js-ai-sdk#192. Go is out of scope for 1.0.

Live check

Passed on 29 Oct: js-ai-sdk#192 at 0a0dd22, Whodunit check --engine langgraph on Bedrock, 15 of 15 graph lines, 0 errors.

Bedrock end-to-end in CI: deferred to Q4 follow-up.

🤖 Generated with Claude Code

---

Note

Overview
Fixes broken LangGraph hand-offs in to_lang_graph (AIC-3592), including the Bedrock failure from passing parallel_tool_calls to bind_tools. Hand-off routing now matches graph(): only the first transfer_to_* call in a turn navigates; extras return an ignored message.

Hand-off Command updates now include a ToolMessage that answers the transfer tool call (via InjectedToolCallId on a small Pydantic _HandoffArgs schema). pydantic>=2.13 is added as a direct dependency. After tools run, a conditional edge stops the routing node from executing again when a hand-off fired, instead of always looping back through _tools → node.

When a node is entered (not continuing its own tool loop), prior conversation is rewritten with _to_neutral_history: reasoning/non-text blocks dropped, tool calls and results flattened to plain [Tool call] / [Tool result] text so downstream agents on other providers never see native tool or reasoning payloads.

Tests add TestNativeHandoffRealStateGraph (six StateGraph scenarios) and TestToNeutralHistory aligned with the TypeScript rewrite fixture.

^{Reviewed by Cursor Bugbot for commit cee3051. Bugbot is set up for automated code reviews on this repo. Configure here.}