feat: added strict-mode fallback for OpenAI tool calls with all_strict

- Added `toolStrictMode` support with `all_strict`/`none`/`mixed` options to OpenAI compatibility.
- Fixed OpenAI-completion strict-mode flows by capturing failed HTTP responses and retrying once as non-strict.
- Fixed completion error reporting by surfacing captured status, headers, and JSON `type`/`param`/`code` details.
- Improved strict-schema enforcement with WeakMap memoization and circular-schema detection in sanitization.
- Fixed OpenRouter provider lookup by resolving fallback model IDs for suffix and date variants in registry resolution.
- Refactored benchmark tooling and added async RPC error-window tracking for scheduled run execution.

This commit is contained in:

can1357

2026-04-13 15:30:34 +02:00

parent b69dea4823

commit 212d56bc11

38 changed files with 1492 additions and 548 deletions

									
										packages/coding-agent/test/bash-executor.test.ts
									
		-1
	
												View File
												
				@@ -116,7 +116,6 @@ describe("executeBash", () => {

						expect(() => process.kill(pid, "SIGKILL")).not.toThrow();

					});

					it("times out commands", async () => {

						if (process.platform === "win32") {

							return;