codebase-brain 0.3.3__tar.gz → 0.3.5__tar.gz

{codebase_brain-0.3.3 → codebase_brain-0.3.5}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: codebase-brain
-Version: 0.3.3
-Summary: An AI brain that lives inside your codebase
+Version: 0.3.5
+Summary: Codebase Brain:Know what you will break and why before you commit it.
 Requires-Python: >=3.8
 Requires-Dist: watchdog
 Requires-Dist: networkx

{codebase_brain-0.3.3 → codebase_brain-0.3.5}/brain_parser/codebase_walker.py

@@ -103,4 +103,4 @@ if __name__ == "__main__":
     save_brain(brain)
 
     loaded = load_brain()
-    print(f"Brain loaded with {len(loaded)} files")
+    print(f"Brain loaded with {len(loaded)} files")

codebase_brain-0.3.5/brain_parser/llm_router.py

@@ -0,0 +1,144 @@
+import os
+from dotenv import load_dotenv
+
+
+def get_config():
+    """Load from local .env first, then global ~/.codebase-brain/config.env"""
+    local_env = os.path.join(os.getcwd(), '.env')
+    if os.path.exists(local_env):
+        load_dotenv(local_env, override=True)
+
+    has_key = os.getenv("LLM_API_KEY") or os.getenv("GROQ_API_KEY")
+
+    if not has_key:
+        global_config = os.path.expanduser("~/.codebase-brain/config.env")
+        if os.path.exists(global_config):
+            load_dotenv(global_config, override=True)
+
+    return {
+        'provider': os.getenv("LLM_PROVIDER", "groq").lower(),
+        'api_key': os.getenv("LLM_API_KEY") or os.getenv("GROQ_API_KEY"),
+        'model': os.getenv("LLM_MODEL") or os.getenv("GROQ_MODEL", "openai/gpt-oss-120b")
+    }
+
+
+def call_llm(prompt, max_tokens=500):
+    config = get_config()
+    provider = config['provider']
+    api_key = config['api_key']
+    model = config['model']
+
+    if provider == "groq":
+        return _call_groq(api_key, model, prompt, max_tokens)
+    elif provider == "openai":
+        return _call_openai(api_key, model, prompt, max_tokens)
+    elif provider == "anthropic":
+        return _call_anthropic(api_key, model, prompt, max_tokens)
+    elif provider == "google":
+        return _call_google(api_key, model, prompt, max_tokens)
+    elif provider == "ollama":
+        return _call_ollama(model, prompt, max_tokens)
+    else:
+        return _call_groq(api_key, model, prompt, max_tokens)
+
+
+def _try_models(model, fallback_list, call_fn):
+    """Generic fallback runner — tries the configured model first, then known-stable alternatives."""
+    models_to_try = [model] + [m for m in fallback_list if m != model]
+    seen = set()
+    models_to_try = [m for m in models_to_try if not (m in seen or seen.add(m))]
+
+    last_error = None
+    for try_model in models_to_try:
+        try:
+            return call_fn(try_model)
+        except Exception as e:
+            last_error = e
+            continue
+
+    return f"Could not get AI response — all models unavailable: {str(last_error)}"
+
+
+def _call_groq(api_key, model, prompt, max_tokens):
+    from groq import Groq
+    client = Groq(api_key=api_key)
+
+    fallback_list = ["openai/gpt-oss-120b", "llama-3.1-8b-instant", "qwen3.6-27b"]
+
+    def call_fn(m):
+        response = client.chat.completions.create(
+            model=m,
+            messages=[{"role": "user", "content": prompt}],
+            max_tokens=max_tokens,
+            temperature=0
+        )
+        return response.choices[0].message.content
+
+    return _try_models(model, fallback_list, call_fn)
+
+
+def _call_openai(api_key, model, prompt, max_tokens):
+    from openai import OpenAI
+    client = OpenAI(api_key=api_key)
+
+    fallback_list = ["gpt-4o-mini", "gpt-4o", "gpt-3.5-turbo"]
+
+    def call_fn(m):
+        response = client.chat.completions.create(
+            model=m,
+            messages=[{"role": "user", "content": prompt}],
+            max_tokens=max_tokens,
+            temperature=0
+        )
+        return response.choices[0].message.content
+
+    return _try_models(model, fallback_list, call_fn)
+
+
+def _call_anthropic(api_key, model, prompt, max_tokens):
+    import anthropic
+    client = anthropic.Anthropic(api_key=api_key)
+
+    fallback_list = ["claude-3-5-haiku-20241022", "claude-3-5-sonnet-20241022", "claude-3-haiku-20240307"]
+
+    def call_fn(m):
+        response = client.messages.create(
+            model=m,
+            max_tokens=max_tokens,
+            messages=[{"role": "user", "content": prompt}]
+        )
+        return response.content[0].text
+
+    return _try_models(model, fallback_list, call_fn)
+
+
+def _call_google(api_key, model, prompt, max_tokens):
+    import google.generativeai as genai
+    genai.configure(api_key=api_key)
+
+    fallback_list = ["gemini-1.5-flash", "gemini-1.5-pro", "gemini-2.0-flash"]
+
+    def call_fn(m):
+        gemini = genai.GenerativeModel(m)
+        response = gemini.generate_content(prompt)
+        return response.text
+
+    return _try_models(model, fallback_list, call_fn)
+
+
+def _call_ollama(model, prompt, max_tokens):
+    import requests
+
+    fallback_list = ["llama3.2", "codellama", "deepseek-coder"]
+
+    def call_fn(m):
+        response = requests.post(
+            "http://localhost:11434/api/generate",
+            json={"model": m, "prompt": prompt, "stream": False}
+        )
+        result = response.json().get("response", "")
+        if not result:
+            raise ValueError(f"Empty response from Ollama model {m}")
+        return result
+
+    return _try_models(model, fallback_list, call_fn)

{codebase_brain-0.3.3 → codebase_brain-0.3.5}/codebase_brain.egg-info/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: codebase-brain
-Version: 0.3.3
-Summary: An AI brain that lives inside your codebase
+Version: 0.3.5
+Summary: Codebase Brain:Know what you will break and why before you commit it.
 Requires-Python: >=3.8
 Requires-Dist: watchdog
 Requires-Dist: networkx

{codebase_brain-0.3.3 → codebase_brain-0.3.5}/setup.py

@@ -2,8 +2,8 @@ from setuptools import setup, find_packages
 
 setup(
     name="codebase-brain",
-    version="0.3.3",
-    description="An AI brain that lives inside your codebase",
+    version="0.3.5",
+    description="Codebase Brain:Know what you will break and why before you commit it.",
     package_dir={"": "."},
     packages=find_packages(where="."),
     install_requires=[

codebase_brain-0.3.3/brain_parser/llm_router.py

@@ -1,96 +0,0 @@
-import os
-from dotenv import load_dotenv
-
-
-def get_config():
-    """Load from local .env first, then global ~/.codebase-brain/config.env"""
-    local_env = os.path.join(os.getcwd(), '.env')
-    if os.path.exists(local_env):
-        load_dotenv(local_env, override=True)
-
-    # check if local .env actually had Brain's keys
-    has_key = os.getenv("LLM_API_KEY") or os.getenv("GROQ_API_KEY")
-
-    if not has_key:
-        # local .env didn't have Brain's config — fall back to global
-        global_config = os.path.expanduser("~/.codebase-brain/config.env")
-        if os.path.exists(global_config):
-            load_dotenv(global_config, override=True)
-
-    return {
-        'provider': os.getenv("LLM_PROVIDER", "groq").lower(),
-        'api_key': os.getenv("LLM_API_KEY") or os.getenv("GROQ_API_KEY"),
-        'model': os.getenv("LLM_MODEL") or os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile")
-    }
-
-
-def call_llm(prompt, max_tokens=500):
-    config = get_config()
-    provider = config['provider']
-    api_key = config['api_key']
-    model = config['model']
-
-    if provider == "groq":
-        return _call_groq(api_key, model, prompt, max_tokens)
-    elif provider == "openai":
-        return _call_openai(api_key, model, prompt, max_tokens)
-    elif provider == "anthropic":
-        return _call_anthropic(api_key, model, prompt, max_tokens)
-    elif provider == "google":
-        return _call_google(api_key, model, prompt, max_tokens)
-    elif provider == "ollama":
-        return _call_ollama(model, prompt, max_tokens)
-    else:
-        return _call_groq(api_key, model, prompt, max_tokens)
-
-
-def _call_groq(api_key, model, prompt, max_tokens):
-    from groq import Groq
-    client = Groq(api_key=api_key)
-    response = client.chat.completions.create(
-        model=model,
-        messages=[{"role": "user", "content": prompt}],
-        max_tokens=max_tokens,
-        temperature=0
-    )
-    return response.choices[0].message.content
-
-
-def _call_openai(api_key, model, prompt, max_tokens):
-    from openai import OpenAI
-    client = OpenAI(api_key=api_key)
-    response = client.chat.completions.create(
-        model=model,
-        messages=[{"role": "user", "content": prompt}],
-        max_tokens=max_tokens,
-        temperature=0
-    )
-    return response.choices[0].message.content
-
-
-def _call_anthropic(api_key, model, prompt, max_tokens):
-    import anthropic
-    client = anthropic.Anthropic(api_key=api_key)
-    response = client.messages.create(
-        model=model,
-        max_tokens=max_tokens,
-        messages=[{"role": "user", "content": prompt}]
-    )
-    return response.content[0].text
-
-
-def _call_google(api_key, model, prompt, max_tokens):
-    import google.generativeai as genai
-    genai.configure(api_key=api_key)
-    gemini = genai.GenerativeModel(model)
-    response = gemini.generate_content(prompt)
-    return response.text
-
-
-def _call_ollama(model, prompt, max_tokens):
-    import requests
-    response = requests.post(
-        "http://localhost:11434/api/generate",
-        json={"model": model, "prompt": prompt, "stream": False}
-    )
-    return response.json().get("response", "")