Stabilize Apex V4: Fixed CLI dependencies, environment loading, and tool execution guards. Enhanced documentation with hyper-logical architecture deep-dives.

Author: Electroiscoding

.github/workflows/ci.yml

@@ -22,30 +22,34 @@ jobs:
           - 6379:6379
 
     steps:
-    - uses: actions/checkout@v4
+      - uses: actions/checkout@v4.2.2
 
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v5
-      with:
-        python-version: ${{ matrix.python-version }}
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5.3.0
+        with:
+          python-version: ${{ matrix.python-version }}
 
-    - name: Install dependencies (including optional xerv)
-      run: |
-        python -m pip install --upgrade pip
-        pip install -e .[dev,xerv]
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e .[dev]
 
-    - name: Run Pytest (Unit + Deep Logic)
-      env:
-        REDIS_URL: redis://localhost:6379
-      run: |
-        pytest tests/ -v --durations=0
+      - name: Run Pytest (Unit + Deep Logic)
+        env:
+          REDIS_URL: redis://localhost:6379
+        run: |
+          pytest tests/ -v --durations=0
         
   benchmark:
     name: "GAIA Regression Test (Level 1)"
     if: github.event_name == 'push' && contains(github.event.head_commit.message, '[bench]')
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v4.2.2
+      - name: Set up Python
+        uses: actions/setup-python@v5.3.0
+        with:
+          python-version: '3.11'
       - name: Install
         run: pip install -e .
       - name: Run Benchmark Script

.github/workflows/main.yml

@@ -15,19 +15,19 @@ jobs:
         python-version: ["3.10", "3.11"]
 
     steps:
-    - uses: actions/checkout@v4
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v5
-      with:
-        python-version: ${{ matrix.python-version }}
-    - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install -e .[dev]
-    - name: Lint with ruff
-      run: |
-        pip install ruff
-        ruff check src/
-    - name: Test with pytest
-      run: |
-        pytest tests/
+      - uses: actions/checkout@v4.2.2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5.3.0
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e .[dev]
+      - name: Lint with ruff
+        run: |
+          pip install ruff
+          ruff check src/
+      - name: Test with pytest
+        run: |
+          pytest tests/

FEATURES_REPORT.md

@@ -0,0 +1,37 @@
+# HANERMA APEX V1.0 - Feature Extraction Report
+
+## 💎 1. Core Orchestration Engine (The "Brain")
+*   **Apex Orchestrator**: Manages multi-agent workflows with stateful reasoning loops.
+*   **Predictive Execution**: Integrates evaluation *during* the loop, not just after.
+*   **Robust Tooling**: Case-insensitive, regex-powered tool call extraction from model text.
+*   **Token-Aware History**: Automatically trims conversation history to prevent context window overflow while preserving key logic.
+*   **Transactional Safety**: Every internal thought and action is committed to an ACID-compliant SQLite bus.
+
+## 🌐 2. Visual Intelligence OS (Apex Dashboard)
+*   **Causal Execution Graph**: Interactive D3.js visualization showing the "why" behind every agent move.
+*   **Interactive Terminal**: A premium user interface for triggering tasks and communicating with agents.
+*   **Telemetry Dashboard**: Live tracking of execution latency (ms), token consumption, and risk scores.
+*   **Historical Log Retrieval**: Instant streaming of all past session logs from the persistent state bus.
+*   **Premium UI/UX**: Custom styling using 'Be Vietnam Pro' and 'Raleway' typography with glassmorphism effects.
+
+## 🛡️ 3. Reliability & Trust Layer
+*   **Predictive Risk Engine**: Scores every prompt for hallucination risk, logic drift, and security threats.
+*   **Symbolic Reasoner**: Performs deterministic checks against a "fact store" to ensure the model isn't inventing data.
+*   **Atomic Guards**: Real-time filters and validators for sensitive operations.
+*   **State Reset capability**: Built-in `/clear` controls to reset the causal bus for fresh experiments.
+
+## 🚀 4. Performance & Hardware Root
+*   **XERV-CRAYON Engine**: High-speed tokenization (15x faster than standard) and vector embedding architecture.
+*   **HCMS (Hyperfast Compressed Memory Store)**: Tiered memory management for long-term semantic retrieval.
+*   **Parallel AST Analyzer**: Optimized analysis of code structures to speed up reasoning over complex codebases.
+
+## 🧩 5. Connectivity & Versatility
+*   **Unified Model Adapter**: A single interface for local (Ollama), cloud (HF, OpenAI), and specialized providers (Together.ai).
+*   **Persona Registry**: Easily spawn agents with specialized identities like `DeepReasoner`, `SystemVerifier`, or `CodeArchitect`.
+*   **Dynamic Tool Binding**: Bind any Python function as an agent tool with zero configuration.
+*   **Self-Healing Routing**: Automatically switches providers or models if a latency or availability threshold is hit.
+
+## 📜 6. Developer Interface
+*   **Minimalist API**: `quick_flow` and `create_agent` patterns for one-line deployment.
+*   **Trace Persistence**: All sessions are saved in `hanerma_state.db` for later analysis or auditing.
+*   **CLI Tools**: Built-in commands for starting the visual server and running demos.