feat: smooth browser challenge harness runs

examples/agent_challenges/browser_click_challenge/README.md

@@ -25,6 +25,10 @@ builds the configured workflow server in the CLI process for each command and
 uses the copied workspace's durable store. It does not reuse in-memory source
 sessions across CLI invocations.
 
+Use `--workspace-template` and `--source-root` to run a same-shape challenge with
+a different prompt template or Python source root. Both default to the bundled
+browser-click example settings.
+
 Use `--start-server` when the trial should exercise the JSON-RPC server path.
 With `--start-server`, the harness starts:
 
@@ -47,6 +51,16 @@ uv run python examples/agent_challenges/browser_click_challenge/run_opencode_tri
   --trials 1
 ```
 
+Variant challenge example:
+
+```powershell
+uv run python examples/agent_challenges/browser_click_challenge/run_opencode_trials.py `
+  --workspace-template examples/agent_challenges/browser_click_challenge/workspace_template `
+  --source-root examples/browser_click_workflow `
+  --workspaces-dir examples/agent_challenges/browser_click_challenge/workspaces_alt `
+  --results-dir examples/agent_challenges/browser_click_challenge/results_alt
+```
+
 Results are written to:
 
 ```text

examples/agent_challenges/browser_click_challenge/run_opencode_trials.py

@@ -235,6 +235,13 @@ def wf_command_prefix_for_config(config_path: Path) -> str:
     return f"uv run wf --config {path_arg} --local"
 
 
+def _display_path(path: Path) -> str:
+    try:
+        return path.resolve().relative_to(ROOT.resolve()).as_posix()
+    except ValueError:
+        return str(path.resolve())
+
+
 def stop_server(process: subprocess.Popen[str]) -> None:
     if process.poll() is not None:
         return
@@ -360,6 +367,18 @@ def main(argv: list[str] | None = None) -> int:
     parser.add_argument("--prompt", type=Path, default=DEFAULT_PROMPT)
     parser.add_argument("--results-dir", type=Path, default=DEFAULT_RESULTS_DIR)
     parser.add_argument("--workspaces-dir", type=Path, default=DEFAULT_WORKSPACES_DIR)
+    parser.add_argument(
+        "--workspace-template",
+        type=Path,
+        default=DEFAULT_WORKSPACE_TEMPLATE,
+        help="Template directory copied for each local-mode trial workspace.",
+    )
+    parser.add_argument(
+        "--source-root",
+        type=Path,
+        default=EXAMPLE_SOURCE_ROOT,
+        help="Python source root written into each generated trial config.",
+    )
     parser.add_argument("--server-url", default=None)
     parser.add_argument("--start-server", action="store_true", default=False)
     parser.add_argument("--no-start-server", action="store_false", dest="start_server")
@@ -407,14 +426,16 @@ def main(argv: list[str] | None = None) -> int:
                     model=args.model,
                     index=index,
                     workspaces_dir=args.workspaces_dir,
+                    template_dir=args.workspace_template,
+                    source_root=args.source_root,
                 )
                 if args.prompt == DEFAULT_PROMPT:
                     prompt_path = workspace.prompt_path
                 trial_wf_command_prefix = wf_command_prefix_for_config(
                     workspace.config_path
                 )
-                workspace_path = workspace.root.relative_to(ROOT).as_posix()
-                config_path = workspace.config_path.relative_to(ROOT).as_posix()
+                workspace_path = _display_path(workspace.root)
+                config_path = _display_path(workspace.config_path)
                 trial_server_context = (
                     "No external workflow RPC server is staged. Use the "
                     "per-trial workspace config copied to "
@@ -437,7 +458,7 @@ def main(argv: list[str] | None = None) -> int:
                     "classification": result["classification"],
                     "returncode": result["returncode"],
                     "duration_seconds": round(float(result["duration_seconds"]), 3),
-                    "report_path": result.get("report_path"),
+                    "report_path": _optional_string(result.get("report_path")),
                     "report_save_error": result.get("report_save_error"),
                 }
             )
@@ -453,5 +474,9 @@ def main(argv: list[str] | None = None) -> int:
             stop_server(managed_server.process)
 
 
+def _optional_string(value: object) -> str | None:
+    return None if value is None else str(value)
+
+
 if __name__ == "__main__":
     raise SystemExit(main())

src/wf_cli/commands/config.py

@@ -5,6 +5,7 @@ from typing import Annotated
 
 import typer
 
+from wf_cli.context import config_path_from_context
 from wf_cli.io import emit_json
 from wf_config import McpSourceConfig, PythonSourceConfig, StdlibSourceConfig
 from wf_config.loader import load_workflow_config
@@ -44,21 +45,25 @@ def migrate_mcp_config(
 
 @app.command("validate")
 def validate_config(
+    ctx: typer.Context,
     config_path: Annotated[
-        Path,
-        typer.Argument(help="Neutral workflow config JSON path."),
-    ],
+        Path | None,
+        typer.Argument(
+            help="Neutral workflow config JSON path. Defaults to global --config.",
+        ),
+    ] = None,
 ) -> None:
     """Validate config shape and trusted static source imports."""
+    resolved_config_path = config_path or Path(config_path_from_context(ctx))
     try:
-        config = load_workflow_config(config_path)
+        config = load_workflow_config(resolved_config_path)
         sources = [_validate_source(source) for source in config.server.sources]
     except Exception as exc:
         raise typer.BadParameter(f"invalid workflow config: {exc}") from exc
     emit_json(
         {
             "valid": True,
-            "path": str(config_path),
+            "path": str(resolved_config_path),
             "sources": sources,
         }
     )

tests/examples/test_opencode_browser_click_challenge.py

@@ -592,6 +592,63 @@ def test_prepare_trial_workspace_uses_next_available_directory(
     assert stale.read_text(encoding="utf-8") == "stale"
 
 
+def test_main_uses_custom_workspace_template_and_source_root(
+    tmp_path: Path,
+    monkeypatch,
+    capsys,
+) -> None:
+    template = tmp_path / "template"
+    template.mkdir()
+    (template / "prompt.md").write_text("custom prompt", encoding="utf-8")
+    source_root = tmp_path / "source"
+    source_root.mkdir()
+    workspaces = tmp_path / "workspaces"
+    results = tmp_path / "results"
+
+    def fake_run_trial(
+        config: TrialConfig,
+        *,
+        index: int,
+        results_dir: Path,
+    ) -> dict[str, object]:
+        return {
+            "index": index,
+            "classification": "success",
+            "returncode": 0,
+            "duration_seconds": 1.0,
+            "report_path": config.prompt_path.parent / "final-report.md",
+        }
+
+    monkeypatch.setattr(run_opencode_trials, "run_trial", fake_run_trial)
+
+    assert (
+        run_opencode_trials.main(
+            [
+                "--model",
+                "check/model",
+                "--trials",
+                "1",
+                "--workspace-template",
+                str(template),
+                "--source-root",
+                str(source_root),
+                "--workspaces-dir",
+                str(workspaces),
+                "--results-dir",
+                str(results),
+            ]
+        )
+        == 0
+    )
+
+    workspace = workspaces / "check_model-trial-001"
+    assert workspace.exists()
+    assert (workspace / "prompt.md").read_text(encoding="utf-8") == "custom prompt"
+    config = json.loads((workspace / "wf.config.json").read_text(encoding="utf-8"))
+    assert config["server"]["sources"][0]["path"] == "../../source"
+    assert '"success_count": 1' in capsys.readouterr().out
+
+
 def test_starting_trial_index_accounts_for_existing_results_and_workspaces(
     tmp_path: Path,
 ) -> None:

tests/wf_cli/test_config_migration.py

@@ -144,6 +144,37 @@ registry = [echo]
     ]
 
 
+def test_wf_config_validate_uses_global_config_when_path_omitted(
+    tmp_path: Path,
+) -> None:
+    config_path = tmp_path / "wf.config.json"
+    config_path.write_text(
+        json.dumps(
+            {
+                "version": 1,
+                "server": {
+                    "store": {"kind": "filesystem", "root": "store"},
+                    "sources": [{"kind": "stdlib", "id": "wf.std"}],
+                },
+            }
+        ),
+        encoding="utf-8",
+    )
+
+    result = CliRunner().invoke(
+        app,
+        ["--config", str(config_path), "config", "validate"],
+    )
+
+    assert result.exit_code == 0, result.output
+    payload = json.loads(result.output)
+    assert payload["valid"] is True
+    assert payload["path"] == str(config_path)
+    assert payload["sources"] == [
+        {"id": "wf.std", "kind": "stdlib", "status": "ok"}
+    ]
+
+
 def test_wf_config_validate_reports_python_source_import_failure(tmp_path: Path) -> None:
     config_path = tmp_path / "wf.config.json"
     config_path.write_text(