agentnova 0.2.4__tar.gz → 0.2.6__tar.gz

{agentnova-0.2.4 → agentnova-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agentnova
-Version: 0.2.4
+Version: 0.2.6
 Summary: A minimal, hackable agentic framework for Ollama and BitNet - local-first AI agent toolkit
 Author-email: VTSTech <veritas@vts-tech.org>
 Maintainer-email: VTSTech <veritas@vts-tech.org>
@@ -34,7 +34,7 @@ Requires-Dist: black>=23.0; extra == "dev"
 Requires-Dist: ruff>=0.1.0; extra == "dev"
 Dynamic: license-file
 
-# ⚛️ AgentNova R02.4
+# ⚛️ AgentNova R02.6
 
 A minimal, hackable agentic framework engineered to run **entirely locally** with [Ollama](https://ollama.com) or [BitNet](https://github.com/microsoft/BitNet).
 
@@ -45,7 +45,7 @@ Inspired by the architecture of OpenClaw, rebuilt from scratch for local-first o
 [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/VTSTech/AgentNova/blob/main/AgentNova.ipynb)
 [![GitHub commits](https://badgen.net/github/commits/VTSTech/AgentNova)](https://GitHub.com/VTSTech/AgentNova/commit/) [![GitHub latest commit](https://badgen.net/github/last-commit/VTSTech/AgentNova)](https://GitHub.com/VTSTech/AgentNova/commit/)
 
-[![pip - agentnova](https://img.shields.io/badge/pip-agentnova-2ea44f?logo=PyPi)](https://pypi.org/project/agentnova/) [![PyPI version fury.io](https://badge.fury.io/py/agentnova.svg)](https://pypi.python.org/pypi/agentnova/) [![PyPI download month](https://img.shields.io/pypi/dm/agentnova.svg)](https://pypi.python.org/pypi/agentnova/) [![PyPI download day](https://img.shields.io/pypi/dd/agentnova.svg)](https://pypi.python.org/pypi/agentnova/)
+[![pip - agentnova](https://img.shields.io/badge/pip-agentnova-2ea44f?logo=PyPi)](https://pypi.org/project/agentnova/) [![PyPI version fury.io](https://badge.fury.io/py/agentnova.svg)](https://pypi.org/project/agentnova/) [![PyPI download month](https://img.shields.io/pypi/dm/agentnova.svg)](https://pypi.org/project/agentnova/) [![PyPI download day](https://img.shields.io/pypi/dd/agentnova.svg)](https://pypi.org/project/agentnova/)
 
 [![License](https://img.shields.io/badge/License-MIT-blue)](#license) [![Go to Python website](https://img.shields.io/badge/dynamic/toml?url=https%3A%2F%2Fraw.githubusercontent.com%2FVTSTech%2FAgentNova%2Frefs%2Fheads%2Fmain%2Fpyproject.toml&query=project.requires-python&label=python&logo=python&logoColor=white)](https://python.org)
 
@@ -201,31 +201,26 @@ agentnova models --tool_support
 
 ### Performance by Tool Support
 
-R02.3 benchmark results (15-test suite):
-
-| Model | Params | Tool Support | Score | Time |
-|-------|--------|--------------|-------|------|
-| **`granite3.1-moe:1b`** | 1B MoE | react | **93% (14/15)** | 87.8s |
-| **`llama3.2:1b`** | 1.2B | native | **87% (13/15)** | 150.2s |
-| `dolphin3.0-qwen2.5:0.5b` | 500M | none | 73% (11/15) | 27.2s |
-| `qwen3:0.6b` | 600M | react | 67% (10/15) | 189.3s |
-| `qwen2.5-coder:0.5b` | 494M | react | 60% (9/15) | 133.1s |
-
-**Quick Diagnostic (Test 15 - 5 questions, ~30s/model):**
-
-| Model | Score | Tool Support |
-|-------|-------|--------------|
-| **`qwen3.5:0.8b`** | **100%** | native |
-| **`qwen2.5:0.5b`** | **100%** | react |
-| `functiongemma:270m` | 80% | native |
-| `granite4:350m` | 80% | native |
-| `qwen3:0.6b` | 60% | react |
-
-**Key improvements in R02.3**:
-- **+13%** for granite3.1-moe:1b (80% → 93%) from few-shot fix
-- **+20%** for llama3.2:1b (67% → 87%) from observation role fix
-- **qwen3:0.6b restored** from 0% (broken) with `think=False` API fix
-- **qwen3.5:0.8b** new sub-1B champion with 100% on quick diagnostic
+R02.6 Quick Diagnostic results (5 questions, ~30-120s/model):
+
+| Model | Score | Time | Tool Support |
+|-------|-------|------|--------------|
+| **`functiongemma:270m`** | **100%** | 19.6s | native |
+| **`granite4:350m`** | **100%** | 49.4s | native |
+| **`qwen2.5:0.5b`** | **100%** | 66.3s | native |
+| **`qwen2.5-coder:0.5b`** | **100%** | 116.5s | react |
+| **`qwen3:0.6b`** | **100%** | 122.8s | react |
+| `gemma3:270m` | 80% | 14.3s | none |
+| `dolphin3.0-qwen2.5:0.5b` | 80% | 26.6s | none |
+| `qwen:0.5b` | 20% | 27.0s | none |
+
+**Key improvements in R02.6**:
+- **5 models achieve 100%** - All tool-calling models now perfect!
+- **functiongemma:270m fastest** at 19.6s (native tools + 270M params)
+- **Multi-step expression extraction** - Handles `8 times 7 minus 5`, word problems, time calculations
+- **ReAct JSON parsing fixed** - Clean extraction even with trailing text
+- **Verbose response fallback** - Uses numeric result when model gives long explanation
+- **Tool-calling outperforms pure reasoning** - ALL native/react models score 100% vs max 80% for no-tool models
 
 ---
 
@@ -276,7 +271,7 @@ Output shows:
 - **Tool Support** - `✓ native`, `ReAct`, `○ none`, or `untested`
 
 ```
-⚛️ AgentNova R02.3 Models
+⚛️ AgentNova R02.6 Models
   Model                                      Family       Context    Tool Support
   ──────────────────────────────────────────────────────────────────────────────
   gemma3:270m                                gemma3       32K        ○ none

{agentnova-0.2.4 → agentnova-0.2.6}/README.md

@@ -1,4 +1,4 @@
-# ⚛️ AgentNova R02.4
+# ⚛️ AgentNova R02.6
 
 A minimal, hackable agentic framework engineered to run **entirely locally** with [Ollama](https://ollama.com) or [BitNet](https://github.com/microsoft/BitNet).
 
@@ -9,7 +9,7 @@ Inspired by the architecture of OpenClaw, rebuilt from scratch for local-first o
 [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/VTSTech/AgentNova/blob/main/AgentNova.ipynb)
 [![GitHub commits](https://badgen.net/github/commits/VTSTech/AgentNova)](https://GitHub.com/VTSTech/AgentNova/commit/) [![GitHub latest commit](https://badgen.net/github/last-commit/VTSTech/AgentNova)](https://GitHub.com/VTSTech/AgentNova/commit/)
 
-[![pip - agentnova](https://img.shields.io/badge/pip-agentnova-2ea44f?logo=PyPi)](https://pypi.org/project/agentnova/) [![PyPI version fury.io](https://badge.fury.io/py/agentnova.svg)](https://pypi.python.org/pypi/agentnova/) [![PyPI download month](https://img.shields.io/pypi/dm/agentnova.svg)](https://pypi.python.org/pypi/agentnova/) [![PyPI download day](https://img.shields.io/pypi/dd/agentnova.svg)](https://pypi.python.org/pypi/agentnova/)
+[![pip - agentnova](https://img.shields.io/badge/pip-agentnova-2ea44f?logo=PyPi)](https://pypi.org/project/agentnova/) [![PyPI version fury.io](https://badge.fury.io/py/agentnova.svg)](https://pypi.org/project/agentnova/) [![PyPI download month](https://img.shields.io/pypi/dm/agentnova.svg)](https://pypi.org/project/agentnova/) [![PyPI download day](https://img.shields.io/pypi/dd/agentnova.svg)](https://pypi.org/project/agentnova/)
 
 [![License](https://img.shields.io/badge/License-MIT-blue)](#license) [![Go to Python website](https://img.shields.io/badge/dynamic/toml?url=https%3A%2F%2Fraw.githubusercontent.com%2FVTSTech%2FAgentNova%2Frefs%2Fheads%2Fmain%2Fpyproject.toml&query=project.requires-python&label=python&logo=python&logoColor=white)](https://python.org)
 
@@ -165,31 +165,26 @@ agentnova models --tool_support
 
 ### Performance by Tool Support
 
-R02.3 benchmark results (15-test suite):
-
-| Model | Params | Tool Support | Score | Time |
-|-------|--------|--------------|-------|------|
-| **`granite3.1-moe:1b`** | 1B MoE | react | **93% (14/15)** | 87.8s |
-| **`llama3.2:1b`** | 1.2B | native | **87% (13/15)** | 150.2s |
-| `dolphin3.0-qwen2.5:0.5b` | 500M | none | 73% (11/15) | 27.2s |
-| `qwen3:0.6b` | 600M | react | 67% (10/15) | 189.3s |
-| `qwen2.5-coder:0.5b` | 494M | react | 60% (9/15) | 133.1s |
-
-**Quick Diagnostic (Test 15 - 5 questions, ~30s/model):**
-
-| Model | Score | Tool Support |
-|-------|-------|--------------|
-| **`qwen3.5:0.8b`** | **100%** | native |
-| **`qwen2.5:0.5b`** | **100%** | react |
-| `functiongemma:270m` | 80% | native |
-| `granite4:350m` | 80% | native |
-| `qwen3:0.6b` | 60% | react |
-
-**Key improvements in R02.3**:
-- **+13%** for granite3.1-moe:1b (80% → 93%) from few-shot fix
-- **+20%** for llama3.2:1b (67% → 87%) from observation role fix
-- **qwen3:0.6b restored** from 0% (broken) with `think=False` API fix
-- **qwen3.5:0.8b** new sub-1B champion with 100% on quick diagnostic
+R02.6 Quick Diagnostic results (5 questions, ~30-120s/model):
+
+| Model | Score | Time | Tool Support |
+|-------|-------|------|--------------|
+| **`functiongemma:270m`** | **100%** | 19.6s | native |
+| **`granite4:350m`** | **100%** | 49.4s | native |
+| **`qwen2.5:0.5b`** | **100%** | 66.3s | native |
+| **`qwen2.5-coder:0.5b`** | **100%** | 116.5s | react |
+| **`qwen3:0.6b`** | **100%** | 122.8s | react |
+| `gemma3:270m` | 80% | 14.3s | none |
+| `dolphin3.0-qwen2.5:0.5b` | 80% | 26.6s | none |
+| `qwen:0.5b` | 20% | 27.0s | none |
+
+**Key improvements in R02.6**:
+- **5 models achieve 100%** - All tool-calling models now perfect!
+- **functiongemma:270m fastest** at 19.6s (native tools + 270M params)
+- **Multi-step expression extraction** - Handles `8 times 7 minus 5`, word problems, time calculations
+- **ReAct JSON parsing fixed** - Clean extraction even with trailing text
+- **Verbose response fallback** - Uses numeric result when model gives long explanation
+- **Tool-calling outperforms pure reasoning** - ALL native/react models score 100% vs max 80% for no-tool models
 
 ---
 
@@ -240,7 +235,7 @@ Output shows:
 - **Tool Support** - `✓ native`, `ReAct`, `○ none`, or `untested`
 
 ```
-⚛️ AgentNova R02.3 Models
+⚛️ AgentNova R02.6 Models
   Model                                      Family       Context    Tool Support
   ──────────────────────────────────────────────────────────────────────────────
   gemma3:270m                                gemma3       32K        ○ none

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/__init__.py

@@ -1,5 +1,5 @@
-"""
-⚛️ AgentNova R02.3 - A minimal, hackable agentic framework for Ollama and BitNet
+"""
+⚛️ AgentNova R02.6 - A minimal, hackable agentic framework for Ollama and BitNet
 
 Written by VTSTech
 https://www.vts-tech.org
@@ -8,7 +8,8 @@ https://github.com/VTSTech/AgentNova
 
 import os
 
-from .core.agent import Agent, AgentRun, StepResult
+from .core.agent import Agent
+from .core.models import AgentRun, StepResult
 from .core.memory import Memory
 from .core.tools import Tool, ToolRegistry, ToolParam
 from .core.ollama_client import OllamaClient
@@ -34,7 +35,7 @@ from .config import (
     AGENTNOVA_BACKEND,
 )
 
-# R00: BitNet backend support
+# R02: BitNet backend support
 try:
     from .bitnet_client import BitnetClient, KNOWN_MODELS
     _BITNET_AVAILABLE = True
@@ -143,7 +144,7 @@ def get_system_prompt(model: str, client=None, default_prompt: str = None):
     
     return default_prompt
 
-# R00 Enhancements
+# R02 Enhancements
 from .core.orchestrator_enhanced import Orchestrator as EnhancedOrchestrator, AgentCard as EnhancedAgentCard
 
 # Tool support detection
@@ -152,7 +153,7 @@ from .cli import get_tool_support
 # Shared args for test scripts
 from .shared_args import add_shared_args, parse_shared_args, SharedConfig
 
-# R00: Agent Mode
+# R02: Agent Mode
 from .agent_mode import (
     AgentMode, AgentState, TaskPlan, Step, Action,
     create_file_write_action, create_file_delete_action,
@@ -167,10 +168,10 @@ __all__ = [
     "OllamaClient", "Orchestrator", "AgentCard",
     # Skills
     "SkillLoader", "Skill", "SkillRegistry",
-    # R00 Enhancements
+    # R02 Enhancements
     "EnhancedOrchestrator", "EnhancedAgentCard",
     "ACPPlugin",
-    # R00: Backend-agnostic helpers
+    # R02: Backend-agnostic helpers
     "get_default_client",
     "get_available_models",
     "get_system_prompt",
@@ -179,12 +180,12 @@ __all__ = [
     "add_shared_args",   # Shared CLI args for test scripts
     "parse_shared_args",
     "SharedConfig",
-    # R00: Agent Mode
+    # R02: Agent Mode
     "AgentMode", "AgentState", "TaskPlan", "Step", "Action",
     "create_file_write_action", "create_file_delete_action",
     "create_mkdir_action", "create_shell_action",
     "format_status", "format_progress",
-    # R02.3: Model Family Configuration
+    # R02.6: Model Family Configuration
     "ModelFamilyConfig", "FAMILY_CONFIGS",
     "get_family_config", "get_stop_tokens", "supports_tools",
     "get_tool_format", "get_preferred_temperature", "should_use_few_shot",
@@ -204,8 +205,8 @@ __all__ = [
 if _BITNET_AVAILABLE:
     __all__.extend(["BitnetClient", "KNOWN_MODELS"])
 
-__version__ = "0.2.4"
+__version__ = "0.2.6"
 __author__ = "VTSTech"
 __author_email__ = "veritas@vts-tech.org"
 __url__ = "https://github.com/VTSTech/AgentNova"
-__website__ = "https://www.vts-tech.org"
+__website__ = "https://www.vts-tech.org"

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/__main__.py

@@ -1,6 +1,6 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
 """
-⚛️ AgentNova R00 — Command Line Interface
+⚛️ AgentNova R02 — Command Line Interface
 
 Entry point for: python -m agentnova [command] [options]
 

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/acp_plugin.py

@@ -1,4 +1,4 @@
-"""
+"""
 ⚛️ AgentNova ACP Plugin - Bridge to Agent Control Panel
 
 This plugin connects AgentNova agents to an ACP (Agent Control Panel) server,
@@ -2350,4 +2350,4 @@ def create_acp_agent(
         on_step=plugin.on_step,
         **agent_kwargs,
     )
-    return agent, plugin
+    return agent, plugin

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/agent_mode.py

@@ -1,6 +1,6 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
 """
-⚛️ AgentNova R00 — Agent Mode
+⚛️ AgentNova R02 — Agent Mode
 
 A goal-driven execution mode where the agent autonomously works through tasks.
 Unlike chat mode (which is user-driven), agent mode:

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/bitnet_client.py

@@ -1,6 +1,6 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
 """
-⚛️ AgentNova R00 — BitnetClient
+⚛️ AgentNova R02 — BitnetClient
 Drop-in replacement for OllamaClient that uses Microsoft's bitnet.cpp
 (llama-server) as the inference backend instead of Ollama.
 
@@ -148,4 +148,4 @@ class BitnetClient:
     
     def supports_streaming(self) -> bool:
         """Return True if this client supports streaming."""
-        return True
+        return True

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/bitnet_setup.py

@@ -1,4 +1,4 @@
-import os
+import os
 import subprocess
 import logging
 
@@ -27,4 +27,4 @@ def setup_bitnet(directory, model_name):
 
 if __name__ == "__main__":
     # Example usage
-    setup_bitnet("F:/Tools/bitnet-llama", "BitNet-b1.58-2B-4T")
+    setup_bitnet("F:/Tools/bitnet-llama", "BitNet-b1.58-2B-4T")

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/cli.py

@@ -1,6 +1,6 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
 """
-⚛️ AgentNova R02.3 — CLI
+⚛️ AgentNova R02.6 — CLI
 Entry point: agentnova <command> [options]
 
 Commands:
@@ -924,7 +924,7 @@ def cmd_models(args):
             print(red("✗  bitnet_client.py not found. Copy it into agentnova/."))
             sys.exit(1)
             
-        print(bold("\n⚛️ AgentNova R02.3 BitNet Model (Remote)"))
+        print(bold("\n⚛️ AgentNova R02.6 BitNet Model (Remote)"))
         
         # We use the client logic to see what is actually running at the URL
         try:
@@ -952,7 +952,7 @@ def cmd_models(args):
         print(yellow("No models found. Pull one with: ollama pull llama3.2:3b"))
         return
 
-    print(bold("\n⚛️ AgentNova R02.3 Models") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold("\n⚛️ AgentNova R02.6 Models") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     
     # Load tested models from storage
     tested_models = _load_tested_models()
@@ -1091,7 +1091,7 @@ def cmd_models(args):
 
 def cmd_tools(args):
     tools = BUILTIN_REGISTRY.all()
-    print(bold("\n⚛️ AgentNova R02.3 Tools") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold("\n⚛️ AgentNova R02.6 Tools") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     print(bold(f"{'Tool':<20} Description"))
     print(dim("─" * 70))
     for t in tools:
@@ -1116,7 +1116,7 @@ def cmd_modelfile(args):
         sys.exit(1)
     
     model = args.model
-    print(bold(f"\n⚛️ AgentNova R02.3 Modelfile") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold(f"\n⚛️ AgentNova R02.6 Modelfile") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     print()
     
     try:
@@ -1181,7 +1181,7 @@ def cmd_skills(args):
     loader = SkillLoader()
     skills = loader.list_skills()
     
-    print(bold("\n⚛️ AgentNova R02.3 Skills") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold("\n⚛️ AgentNova R02.6 Skills") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     
     if not skills:
         print(yellow("  No skills found."))
@@ -1266,7 +1266,7 @@ def cmd_run(args):
     if acp_plugin:
         bootstrap_result = acp_plugin.bootstrap(claim_primary=False)  # AgentNova is secondary
 
-    print(bold("⚛️ AgentNova R02.3") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold("⚛️ AgentNova R02.6") + dim(" · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     print(f"Prompt: {args.prompt}")
 
     # Log user message to ACP
@@ -1359,7 +1359,7 @@ def cmd_chat(args):
     parts.append("]")
     status = " ".join(parts)
     
-    print(bold(f"\n⚛️ AgentNova R02.3 chat") + dim(f"  {status} · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold(f"\n⚛️ AgentNova R02.6 chat") + dim(f"  {status} · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     print(dim("  Type 'exit', 'quit', or Ctrl+C to quit."))
     print(dim("  Type '/help' to see all available commands."))
     print(dim("  Type '/ollama' to manage Ollama models (works with remote Ollama)."))
@@ -1909,7 +1909,7 @@ def cmd_agent(args):
     parts.append("]")
     status = " ".join(parts)
 
-    print(bold(f"\n⚛️ AgentNova R02.3 agent mode") + dim(f"  {status} · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
+    print(bold(f"\n⚛️ AgentNova R02.6 agent mode") + dim(f"  {status} · Written by VTSTech · https://www.vts-tech.org · https://github.com/VTSTech/AgentNova"))
     print(dim("  Agent mode: Give tasks and the agent will work autonomously."))
     print(dim("  Type '/help' to see available commands."))
     print(dim("  ─────────────────────────────────────"))
@@ -2291,11 +2291,14 @@ def cmd_test(args):
         "14": "GSM8K benchmark - 50 math questions",
         "14_acp": "GSM8K benchmark with ACP tracking",
         "15": "Quick diagnostic - 5 questions, ~30s/model",
+        "15_acp": "Quick diagnostic with ACP tracking",
+        "16": "Agent Mode test - 7 autonomous task tests",
+        "16_acp": "Agent Mode test with ACP tracking",
         "backend": "Backend demo - Ollama/BitNet switching",
     }
     
     # Quick test set (skips long-running benchmarks)
-    QUICK_TESTS = ["01", "02", "03", "04", "05", "10", "11", "15"]
+    QUICK_TESTS = ["01", "02", "03", "04", "05", "10", "11", "15", "16"]
     
     if args.list:
         print(bold("\n⚛️ AgentNova Test Examples"))
@@ -2457,7 +2460,7 @@ def build_parser() -> argparse.ArgumentParser:
 
     parser = argparse.ArgumentParser(
         prog="agentnova",
-        description="⚛️ AgentNova R02.3 - local agentic AI powered by Ollama",
+        description="⚛️ AgentNova R02.6 - local agentic AI powered by Ollama",
     )
 
     sub = parser.add_subparsers(dest="command", metavar="command")

{agentnova-0.2.4 → agentnova-0.2.6}/agentnova/config.py

@@ -1,5 +1,5 @@
-"""
-⚛️ AgentNova R00 — Central Configuration
+"""
+⚛️ AgentNova R02 — Central Configuration
 
 Single source of truth for Ollama and ACP server URLs.
 Change these values once to update all tests and examples.