codebind 0.4.2__tar.gz → 0.4.3__tar.gz

{codebind-0.4.2 → codebind-0.4.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codebind
-Version: 0.4.2
+Version: 0.4.3
 Summary: A frontend-neutral model loop for persistent IPython sessions.
 Keywords: ai,ipython,llm,repl,agents
 Author: Giovanni Gravili
@@ -47,11 +47,9 @@ After installation, start ordinary IPython with the Codebind extension from any
 codebind
 ```
 
-It opens standard IPython with `chat` and `Models` in the user namespace. All normal IPython command-line options remain available.
+It opens standard IPython with `chat`, `models`, and `Models` in the user namespace. `models` loads provider values from `$XDG_CONFIG_HOME/codebind/models.json`, or `~/.config/codebind/models.json` when `XDG_CONFIG_HOME` is unset. Keep that credential file readable only by its owner. All normal IPython command-line options remain available.
 
 ```python
-models = Models({"openai": "OPENAI_API_KEY"})
-
 chat.send(
     "Inspect this project and tell me what to implement first.",
     models.chat("openai/gpt-5", reasoning_effort="medium"),
@@ -65,7 +63,7 @@ Codebind does not load files or construct a project prompt automatically. The us
 Start JupyterLab with Codebind from any directory without a permanent installation:
 
 ```console
-uvx --from jupyterlab --with codebind jupyter lab
+uvx --from jupyterlab --with codebind --with 'nbconvert[webpdf]' jupyter lab
 ```
 
 Or install both packages into the same environment, then start JupyterLab normally:
@@ -84,7 +82,6 @@ Load Codebind in a notebook:
 Run that cell once so JupyterLab can connect to the extension, then use Codebind in later cells:
 
 ```python
-models = Models({"openai": "OPENAI_API_KEY"})
 model = models.chat("openai/gpt-5")
 await chat.asend("Inspect the current notebook state.", model)
 ```
@@ -111,6 +108,6 @@ chat.send(
 )
 ```
 
-The authorization remains in memory for this session. Persistent credential storage belongs to the host application.
+The authorization remains in memory for this session. Save provider values as a JSON object in Codebind's XDG configuration file and reload the extension to expose them through `models`.
 
 OpenAI officially supports ChatGPT subscription sign-in for Codex clients. Models Provider reproduces that account-access boundary for this library; it is separate from the public, pay-as-you-go OpenAI API and may require compatibility updates when the Codex account protocol changes.

{codebind-0.4.2 → codebind-0.4.3}/README.md

@@ -22,11 +22,9 @@ After installation, start ordinary IPython with the Codebind extension from any
 codebind
 ```
 
-It opens standard IPython with `chat` and `Models` in the user namespace. All normal IPython command-line options remain available.
+It opens standard IPython with `chat`, `models`, and `Models` in the user namespace. `models` loads provider values from `$XDG_CONFIG_HOME/codebind/models.json`, or `~/.config/codebind/models.json` when `XDG_CONFIG_HOME` is unset. Keep that credential file readable only by its owner. All normal IPython command-line options remain available.
 
 ```python
-models = Models({"openai": "OPENAI_API_KEY"})
-
 chat.send(
     "Inspect this project and tell me what to implement first.",
     models.chat("openai/gpt-5", reasoning_effort="medium"),
@@ -40,7 +38,7 @@ Codebind does not load files or construct a project prompt automatically. The us
 Start JupyterLab with Codebind from any directory without a permanent installation:
 
 ```console
-uvx --from jupyterlab --with codebind jupyter lab
+uvx --from jupyterlab --with codebind --with 'nbconvert[webpdf]' jupyter lab
 ```
 
 Or install both packages into the same environment, then start JupyterLab normally:
@@ -59,7 +57,6 @@ Load Codebind in a notebook:
 Run that cell once so JupyterLab can connect to the extension, then use Codebind in later cells:
 
 ```python
-models = Models({"openai": "OPENAI_API_KEY"})
 model = models.chat("openai/gpt-5")
 await chat.asend("Inspect the current notebook state.", model)
 ```
@@ -86,6 +83,6 @@ chat.send(
 )
 ```
 
-The authorization remains in memory for this session. Persistent credential storage belongs to the host application.
+The authorization remains in memory for this session. Save provider values as a JSON object in Codebind's XDG configuration file and reload the extension to expose them through `models`.
 
 OpenAI officially supports ChatGPT subscription sign-in for Codex clients. Models Provider reproduces that account-access boundary for this library; it is separate from the public, pay-as-you-go OpenAI API and may require compatibility updates when the Codex account protocol changes.

{codebind-0.4.2 → codebind-0.4.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "codebind"
-version = "0.4.2"
+version = "0.4.3"
 description = "A frontend-neutral model loop for persistent IPython sessions."
 readme = "README.md"
 requires-python = ">=3.13"

{codebind-0.4.2 → codebind-0.4.3}/pyproject.toml.orig

@@ -1,6 +1,6 @@
 [project]
 name = "codebind"
-version = "0.4.2"
+version = "0.4.3"
 description = "A frontend-neutral model loop for persistent IPython sessions."
 readme = "README.md"
 authors = [

{codebind-0.4.2 → codebind-0.4.3}/src/codebind/extension.py

@@ -2,6 +2,9 @@
 
 from __future__ import annotations
 
+import json
+import os
+from pathlib import Path
 from typing import Any
 
 from IPython.core.interactiveshell import InteractiveShell
@@ -14,8 +17,28 @@ from .session import Session
 _NAMESPACE_ATTRIBUTE = "_codebind_extension_namespace"
 
 
+def _models_path() -> Path:
+    configured = os.environ.get("XDG_CONFIG_HOME")
+    if configured:
+        config_home = Path(configured).expanduser()
+        if config_home.is_absolute():
+            return config_home / "codebind" / "models.json"
+    return Path.home() / ".config" / "codebind" / "models.json"
+
+
+def _load_models() -> Models:
+    path = _models_path()
+    if not path.exists():
+        return Models()
+    values = json.loads(path.read_text())
+    if not isinstance(values, dict):
+        raise ValueError(f"Codebind model configuration must be a JSON object: {path}")
+    return Models(values)
+
+
 def load_ipython_extension(ipython: InteractiveShell) -> None:
     """Load Codebind into the active IPython user namespace."""
+    models = _load_models()
     previous = getattr(ipython, _NAMESPACE_ATTRIBUTE, None)
     if isinstance(previous, dict):
         previous_chat = previous.get("chat")
@@ -26,6 +49,7 @@ def load_ipython_extension(ipython: InteractiveShell) -> None:
     namespace: dict[str, Any] = {
         "chat": Session(shell=ipython, bridge=bridge),
         "Models": Models,
+        "models": models,
     }
     ipython.push(namespace)
     setattr(ipython, _NAMESPACE_ATTRIBUTE, namespace)