Abigail/embedding OpenAI base url (#1068)

* fix(cli): write embedding base url in --setup

* feat(cli): surface local stack on the welcome screen

honcho-cli/CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 
 ## [Unreleased]
 
+### Fixed
+
+- `--setup` for openai-compatible writes `EMBEDDING_MODEL_CONFIG__OVERRIDES__BASE_URL` into the profile `.env` alongside `LLM_OPENAI_BASE_URL`
+
 ## [0.1.3] - 2026-08-25
 
 ### Added

honcho-cli/src/honcho_cli/_help.py

@@ -59,16 +59,21 @@ def _welcome_panel(title: str, rows: list[tuple[str, str]]) -> Panel:
 
 
 def print_welcome(console: Console) -> None:
-    """Render the curated 3-panel welcome (banner + getting started / memory / commands)."""
+    """Render the curated welcome (banner + getting started / local stack / commands / memory)."""
     if use_json():
         return
     console.print(f"[bold {BRAND}]{BANNER}[/bold {BRAND}]")
     console.print(f"  [dim]v{__version__}[/dim]\n", highlight=False)
 
     start_rows = [
-        ("honcho init",   "configure API key and server URL"),
-        ("honcho start",  "run a local Honcho stack (Docker)"),
-        ("honcho doctor", "verify connection and workspace health"),
+        ("honcho init",                  "configure API key and server URL"),
+        ("honcho start [--setup basic]", "run a local Honcho stack (Docker)"),
+        ("honcho doctor",                "verify connection and workspace health"),
+    ]
+    stack_rows = [
+        ("honcho start / status / stop",          "lifecycle for the local Docker stack"),
+        ("honcho start --setup basic",            "interactive LLM + feature wizard"),
+        ("HONCHO_BASE_URL=http://127.0.0.1:8000", "prefix any command — CLI stays on api.honcho.dev until you set this"),
     ]
     cmd_rows = [
         ("[dim]pattern[/dim]", r"[dim]honcho <command> \[args] \[-w workspace] \[-p peer] \[-s session][/dim]"),
@@ -85,14 +90,15 @@ def print_welcome(console: Console) -> None:
         ("config",     "inspect current configuration"),
     ]
     memory_rows = [
-        ("honcho peer chat \"...\" -p <peer> -w <workspace>","query the Dialectic about a peer"),
-        ("honcho peer inspect -p <peer> -w <workspace>","dashboard: peer card + recent conclusions + configuration"),
+        ("honcho peer chat \"...\" -p <peer> -w <workspace>", "query the Dialectic about a peer"),
+        ("honcho peer inspect -p <peer> -w <workspace>", "dashboard: peer card + recent conclusions + configuration"),
         ("honcho peer representation -p <peer> -w <workspace>", "global peer representation"),
         ("honcho peer representation -p <peer> -w <workspace> -s <session>", "session-scoped peer representation"),
         ("honcho peer card -p <peer> -w <workspace>", "synthesized identity: traits, preferences, instructions"),
-        ("honcho conclusion list -p <peer> -w <workspace>",  "browse peer conclusions"),
+        ("honcho conclusion list -p <peer> -w <workspace>", "browse peer conclusions"),
+        ("honcho session view / context -s <session>", "transcript, or what an agent would see"),
+        ("honcho workspace queue-status", "is the deriver processing?"),
     ]
-
     option_rows = [
         ("-w / --workspace", "scope to a workspace"),
         ("-p / --peer",      "scope to a peer"),
@@ -102,6 +108,7 @@ def print_welcome(console: Console) -> None:
     ]
 
     console.print(_welcome_panel("getting started", start_rows))
+    console.print(_welcome_panel("local stack", stack_rows))
     console.print(_welcome_panel("commands", cmd_rows))
     console.print(_welcome_panel("memory", memory_rows))
     console.print(_welcome_panel("options", option_rows))

honcho-cli/src/honcho_cli/local/setup.py

@@ -14,11 +14,7 @@ from pathlib import Path
 import typer
 from rich.console import Console
 
-from honcho_cli.local.env import (
-    is_placeholder_key,
-    read_env_file,
-    settings_from_environ,
-)
+from honcho_cli.local.env import is_placeholder_key, read_env_file, settings_from_environ
 from honcho_cli.output import print_error
 
 SETUP_MODES = ("basic", "advanced")
@@ -145,6 +141,10 @@ def answers_to_env(answers: SetupAnswers) -> dict[str, str]:
     env[_PROVIDER_KEY_ENV[answers.provider]] = answers.api_key
     if answers.base_url:
         env["LLM_OPENAI_BASE_URL"] = answers.base_url
+        # Embeddings do not inherit this URL; write it so OpenRouter/vLLM
+        # keys are not sent to api.openai.com.
+        if (answers.embedding_transport or "openai") == "openai":
+            env["EMBEDDING_MODEL_CONFIG__OVERRIDES__BASE_URL"] = answers.base_url
 
     if answers.embedding_api_key and answers.embedding_key_transport:
         embed_key = (
@@ -188,9 +188,13 @@ def answers_to_env(answers: SetupAnswers) -> dict[str, str]:
 
 def answers_drop_keys(answers: SetupAnswers) -> tuple[str, ...]:
     """Keys to remove so a previous wizard run cannot leak into this one."""
-    if answers.provider == "openai-compatible":
-        return ()
-    return ("LLM_OPENAI_BASE_URL",)
+    drop: list[str] = []
+    if answers.provider != "openai-compatible":
+        drop.append("LLM_OPENAI_BASE_URL")
+    embed_openai = (answers.embedding_transport or "openai") == "openai"
+    if answers.provider != "openai-compatible" or not embed_openai:
+        drop.append("EMBEDDING_MODEL_CONFIG__OVERRIDES__BASE_URL")
+    return tuple(drop)
 
 
 def run_setup(

honcho-cli/tests/test_setup.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 from honcho_cli.local.setup import (
     DIALECTIC_LEVELS,
     SetupAnswers,
+    answers_drop_keys,
     answers_to_env,
     chat_model_default,
     load_toml_setup_defaults,
@@ -46,6 +47,33 @@ def test_basic_anthropic_keeps_openai_embeddings_default():
     assert "EMBEDDING_MODEL_CONFIG__TRANSPORT" not in env
 
 
+def test_openai_compatible_copies_base_url_to_embeddings():
+    env = answers_to_env(
+        SetupAnswers(
+            mode="basic",
+            provider="openai-compatible",
+            api_key="sk-or-test",
+            chat_model="gpt-test",
+            base_url="https://openrouter.ai/api/v1",
+        )
+    )
+    assert env["LLM_OPENAI_BASE_URL"] == "https://openrouter.ai/api/v1"
+    assert (
+        env["EMBEDDING_MODEL_CONFIG__OVERRIDES__BASE_URL"]
+        == "https://openrouter.ai/api/v1"
+    )
+
+
+def test_leaving_openai_compatible_drops_proxy_urls():
+    dropped = answers_drop_keys(
+        SetupAnswers(
+            mode="basic", provider="openai", api_key="sk", chat_model="gpt-test"
+        )
+    )
+    assert "LLM_OPENAI_BASE_URL" in dropped
+    assert "EMBEDDING_MODEL_CONFIG__OVERRIDES__BASE_URL" in dropped
+
+
 def test_chat_default_comes_from_image_toml(tmp_path):
     path = tmp_path / "config.toml"
     path.write_text(