saksh 1.0.0__tar.gz

saksh-1.0.0/MANIFEST.in

@@ -0,0 +1 @@
+recursive-include saksh/dashboard/static *

saksh-1.0.0/PKG-INFO

@@ -0,0 +1,118 @@
+Metadata-Version: 2.4
+Name: saksh
+Version: 1.0.0
+Summary: Sahasraksh - The All-Seeing Observer for AI Agents
+Author-email: Ram Bikkina <itsrambikkina@gmail.com>
+Keywords: observability,ai-agents,crewai,langgraph
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: fastapi
+Requires-Dist: uvicorn
+Requires-Dist: requests
+Requires-Dist: pydantic
+Requires-Dist: crewai
+Requires-Dist: build>=1.3.0
+Requires-Dist: twine>=6.0.1
+
+# 👁️ Saksh (Sahasraksh) v1.0.0
+
+> **"The All-Seeing Observer for your AI Agents."**
+
+**Saksh** (derived from *Sahasraksh* — "The One with a Thousand Eyes") is a lightweight, non-intrusive observability layer designed specifically for modern AI Agent ecosystems.
+
+Think of it as a **Sidecar** for your agents. whether you're running a swarm of **CrewAI** workers, a complex **LangGraph** state machine, or raw LLM chains, Saksh sits quietly in the background, monitoring health, latency, and vitality without ever blocking your main event loop.
+
+![Saksh Dashboard](./saksh/dashboard/static/image.png)
+
+
+## ⚡ Why Saksh?
+
+Building agents is hard. Debugging why they "hung" or "timed out" is harder.
+Saksh gives you a **Mission Control** dashboard out of the box, so you can stop guessing if your agents are actually working or just hallucinating silence.
+
+* **Universal & Agnostic:** We don't care if you use CrewAI, LangGraph, or raw OpenAI calls. If it has an API key and a URL, we can watch it.
+* **Zero-Touch Dashboard:** Just instantiate the class, and a full React/FastAPI dashboard spins up on port `2604`. No React code required.
+* **Vitality Scoring:** We don't just check "Is it up?". We calculate a **Vitality Score (0-100)** based on latency spikes and auth failures.
+* **Non-Blocking Sidecar:** Runs in a daemon thread. Your agents keep working even if Saksh takes a coffee break.
+
+## 📦 Installation
+
+Get it via pip:
+
+```bash
+pip install saksh
+
+```
+
+## 🛠️ Quick Start
+
+You can attach Saksh to your existing stack in about 3 lines of code.
+
+### 1. The Setup
+
+```python
+from saksh import Saksh, CrewAINetra, LangGraphNetra
+from crewai import Agent
+
+# 1. Summon the Observer 👁️
+# This auto-magically starts the dashboard at http://localhost:2604
+observer = Saksh(start_dashboard=True)
+
+```
+
+### 2. Connect CrewAI
+
+Saksh automatically detects the LLM configuration (URL, API Key) inside your CrewAI agents.
+
+```python
+# Your standard CrewAI setup
+researcher = Agent(
+    role="Researcher", 
+    goal="Analyze market trends", 
+    backstory="You are a data wizard."
+)
+
+# Open an Eye on it
+observer.open_eye(CrewAINetra(researcher, "Market-Researcher-01"))
+
+```
+
+### 3. Connect LangGraph
+
+Since LangGraph is stateful and abstract, we monitor the underlying LLM provider it relies on.
+
+```python
+# graph_app = workflow.compile()
+
+# Register the graph (defaults to checking OpenAI availability)
+observer.open_eye(LangGraphNetra(graph_app, "Math-Graph-State-Machine"))
+
+```
+
+*That's it. Your terminal will now log:*
+`👁️ Saksh is watching...`
+
+## 🧩 Architecture
+
+Saksh uses a **Netra (Eye) Adapter Pattern**.
+
+* **Saksh (Core):** The singleton observer that manages the background thread and dashboard.
+* **Netra (Adapter):** A standardized interface that knows how to "Gaze" at a specific type of agent and normalize its health data into a `HealthPulse`.
+
+## 🤝 Contributing
+
+This is **v1.0.0** — the "It Works on My Machine" release.
+We are actively looking for contributors to build **Netras** (Adapters) for:
+
+* Autogen
+* Semantic Kernel
+* Local Ollama/Llama.cpp instances
+
+PRs are welcome! Let's build the standard for Agent Observability together.
+
+## 📜 License
+
+MIT © 2026 Ram Bikkina

saksh-1.0.0/README.md

@@ -0,0 +1,99 @@
+# 👁️ Saksh (Sahasraksh) v1.0.0
+
+> **"The All-Seeing Observer for your AI Agents."**
+
+**Saksh** (derived from *Sahasraksh* — "The One with a Thousand Eyes") is a lightweight, non-intrusive observability layer designed specifically for modern AI Agent ecosystems.
+
+Think of it as a **Sidecar** for your agents. whether you're running a swarm of **CrewAI** workers, a complex **LangGraph** state machine, or raw LLM chains, Saksh sits quietly in the background, monitoring health, latency, and vitality without ever blocking your main event loop.
+
+![Saksh Dashboard](./saksh/dashboard/static/image.png)
+
+
+## ⚡ Why Saksh?
+
+Building agents is hard. Debugging why they "hung" or "timed out" is harder.
+Saksh gives you a **Mission Control** dashboard out of the box, so you can stop guessing if your agents are actually working or just hallucinating silence.
+
+* **Universal & Agnostic:** We don't care if you use CrewAI, LangGraph, or raw OpenAI calls. If it has an API key and a URL, we can watch it.
+* **Zero-Touch Dashboard:** Just instantiate the class, and a full React/FastAPI dashboard spins up on port `2604`. No React code required.
+* **Vitality Scoring:** We don't just check "Is it up?". We calculate a **Vitality Score (0-100)** based on latency spikes and auth failures.
+* **Non-Blocking Sidecar:** Runs in a daemon thread. Your agents keep working even if Saksh takes a coffee break.
+
+## 📦 Installation
+
+Get it via pip:
+
+```bash
+pip install saksh
+
+```
+
+## 🛠️ Quick Start
+
+You can attach Saksh to your existing stack in about 3 lines of code.
+
+### 1. The Setup
+
+```python
+from saksh import Saksh, CrewAINetra, LangGraphNetra
+from crewai import Agent
+
+# 1. Summon the Observer 👁️
+# This auto-magically starts the dashboard at http://localhost:2604
+observer = Saksh(start_dashboard=True)
+
+```
+
+### 2. Connect CrewAI
+
+Saksh automatically detects the LLM configuration (URL, API Key) inside your CrewAI agents.
+
+```python
+# Your standard CrewAI setup
+researcher = Agent(
+    role="Researcher", 
+    goal="Analyze market trends", 
+    backstory="You are a data wizard."
+)
+
+# Open an Eye on it
+observer.open_eye(CrewAINetra(researcher, "Market-Researcher-01"))
+
+```
+
+### 3. Connect LangGraph
+
+Since LangGraph is stateful and abstract, we monitor the underlying LLM provider it relies on.
+
+```python
+# graph_app = workflow.compile()
+
+# Register the graph (defaults to checking OpenAI availability)
+observer.open_eye(LangGraphNetra(graph_app, "Math-Graph-State-Machine"))
+
+```
+
+*That's it. Your terminal will now log:*
+`👁️ Saksh is watching...`
+
+## 🧩 Architecture
+
+Saksh uses a **Netra (Eye) Adapter Pattern**.
+
+* **Saksh (Core):** The singleton observer that manages the background thread and dashboard.
+* **Netra (Adapter):** A standardized interface that knows how to "Gaze" at a specific type of agent and normalize its health data into a `HealthPulse`.
+
+## 🤝 Contributing
+
+This is **v1.0.0** — the "It Works on My Machine" release.
+We are actively looking for contributors to build **Netras** (Adapters) for:
+
+* Autogen
+* Semantic Kernel
+* Local Ollama/Llama.cpp instances
+
+PRs are welcome! Let's build the standard for Agent Observability together.
+
+## 📜 License
+
+MIT © 2026 Ram Bikkina

saksh-1.0.0/pyproject.toml

@@ -0,0 +1,31 @@
+[build-system]
+requires = ["setuptools", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "saksh"
+version = "1.0.0"
+description = "Sahasraksh - The All-Seeing Observer for AI Agents"
+readme = "README.md" 
+authors = [{name = "Ram Bikkina", email = "itsrambikkina@gmail.com"}]
+dependencies = [
+    "fastapi",
+    "uvicorn",
+    "requests",
+    "pydantic",
+    "crewai",
+    "build>=1.3.0",
+    "twine>=6.0.1",
+]
+requires-python = ">=3.9"
+
+# Optional: Add keywords and classifiers to help people find your library
+keywords = ["observability", "ai-agents", "crewai", "langgraph"]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+]
+
+[tool.setuptools.package-data]
+saksh = ["dashboard/static/*"]

saksh-1.0.0/saksh/__init__.py

@@ -0,0 +1,4 @@
+from .core import Saksh, Netra, HealthPulse
+from .adapters import CrewAINetra, LangGraphNetra
+
+__all__ = ["Saksh", "Netra", "HealthPulse", "CrewAINetra", "LangGraphNetra"]

saksh-1.0.0/saksh/adapters.py

@@ -0,0 +1,136 @@
+import requests
+import time
+import os
+from typing import Optional, Dict, Any
+from .core import Netra, HealthPulse
+
+# --- SHARED HELPER FUNCTION ---
+def _perform_health_check(agent_id: str, url: str, headers: Dict[str, str], timeout: int = 5) -> HealthPulse:
+    """
+    Internal helper to ping a URL and return a standardized HealthPulse.
+    Used by both CrewAI and LangGraph adapters to avoid code duplication.
+    """
+    start = time.time()
+    status = "down"
+    score = 0
+    
+    try:
+        # Perform the actual network request
+        response = requests.get(url, headers=headers, timeout=timeout)
+        
+        # --- STRICT STATUS CHECK ---
+        # 200 = Healthy (Key worked, Server up)
+        # 401 = DOWN (Key Failed - Critical Logic)
+        # 404 = Degraded (Server up, endpoint might be slightly wrong but alive)
+        
+        if response.status_code == 200:
+            status = "healthy"
+            score = 100
+        elif response.status_code == 401:
+            status = "down" 
+            score = 0
+            print(f"⚠️ SAKSH WARNING: Agent {agent_id} Failed Auth (401)")
+        else:
+            status = "degraded"
+            score = 50
+            
+    except requests.exceptions.Timeout:
+        status = "degraded" # It's alive but slow
+        score = 40
+        print(f"⚠️ SAKSH WARNING: Agent {agent_id} Timed Out")
+    except Exception:
+        status = "down"
+        score = 0
+        
+    # Calculate Latency
+    latency = (time.time() - start) * 1000
+    
+    # Adjust Score based on Latency (Penalize Slowness)
+    if status == "healthy":
+        if latency > 1500: score -= 20
+        if latency > 3000: score -= 40
+        
+    return HealthPulse(
+        agent_id=agent_id,
+        status=status,
+        latency_ms=round(latency, 2),
+        vitality_score=max(0, score) # Ensure score never goes below 0
+    )
+
+
+# --- 1. CREWAI ADAPTER ---
+class CrewAINetra(Netra):
+    """
+    Specialized for CrewAI Agents.
+    Automatically extracts: agent.llm.base_url and agent.llm.api_key
+    """
+    def __init__(self, crewai_agent, agent_id: str):
+        super().__init__(agent_id)
+        self.agent = crewai_agent
+        
+        # 1. Logic to extract URL
+        # CrewAI often hides the base_url inside the llm object
+        if hasattr(crewai_agent.llm, 'base_url') and crewai_agent.llm.base_url:
+            self.target_url = crewai_agent.llm.base_url
+            self.is_custom = True
+        else:
+            self.target_url = "https://api.openai.com/v1"
+            self.is_custom = False
+            
+        # 2. Logic to extract API Key
+        self.api_key = getattr(crewai_agent.llm, 'api_key', None)
+        
+        # Fallback: Check Environment Variable if key is missing in object
+        if not self.api_key and not self.is_custom:
+             self.api_key = os.getenv("OPENAI_API_KEY")
+
+    def gaze(self) -> HealthPulse:
+        # Construct the Check URL
+        if self.is_custom:
+            base = self.target_url.rstrip("/")
+            if base.endswith("/v1"):
+                check_url = f"{base}/models"
+            else:
+                check_url = f"{base}/v1/models"
+        else:
+            check_url = "https://api.openai.com/v1/models"
+
+        # Construct Headers
+        headers = {}
+        if self.api_key and self.api_key != "na":
+            headers["Authorization"] = f"Bearer {self.api_key}"
+
+        # Delegate to shared helper
+        return _perform_health_check(self.agent_id, check_url, headers)
+
+
+# --- 2. LANGGRAPH ADAPTER ---
+class LangGraphNetra(Netra):
+    """
+    Specialized for LangGraph Compiled Graphs.
+    Since Graphs are abstract, we monitor the underlying LLM provider they rely on.
+    """
+    def __init__(self, graph_app, agent_id: str, target_url="https://api.openai.com/v1", api_key=None):
+        super().__init__(agent_id)
+        self.graph = graph_app
+        
+        # LangGraph allows flexible backends, so we accept overrides.
+        # Defaults to OpenAI if not specified.
+        self.target_url = target_url
+        self.api_key = api_key or os.getenv("OPENAI_API_KEY")
+
+    def gaze(self) -> HealthPulse:
+        # Construct URL
+        base = self.target_url.rstrip("/")
+        if base.endswith("/v1"):
+            check_url = f"{base}/models"
+        else:
+            check_url = f"{base}/v1/models"
+            
+        # Construct Headers
+        headers = {}
+        if self.api_key:
+            headers["Authorization"] = f"Bearer {self.api_key}"
+
+        # Delegate to shared helper
+        return _perform_health_check(self.agent_id, check_url, headers)

saksh-1.0.0/saksh/core.py

@@ -0,0 +1,92 @@
+import threading
+import uvicorn
+import os
+import sys
+from typing import List, Dict, Any, Literal
+from datetime import datetime
+from pydantic import BaseModel, Field
+import abc
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.staticfiles import StaticFiles
+
+# --- SCHEMAS ---
+class HealthPulse(BaseModel):
+    agent_id: str
+    status: Literal["healthy", "degraded", "down"]
+    latency_ms: float
+    timestamp: datetime = Field(default_factory=datetime.now)
+    vitality_score: int 
+
+class Netra(abc.ABC):
+    def __init__(self, agent_id: str):
+        self.agent_id = agent_id
+
+    @abc.abstractmethod
+    def gaze(self) -> HealthPulse:
+        pass
+
+class Saksh:
+    _instance = None 
+
+    def __new__(cls, *args, **kwargs):
+        if not cls._instance:
+            cls._instance = super(Saksh, cls).__new__(cls)
+        return cls._instance
+
+    def __init__(self, start_dashboard=True, port=2604):
+        if hasattr(self, '_initialized'):
+            return
+            
+        self._thousand_eyes: List[Netra] = []
+        self._initialized = True
+        
+        if start_dashboard:
+            self._start_dashboard_thread(port)
+
+    def open_eye(self, adapter: Netra):
+        self._thousand_eyes.append(adapter)
+
+    def observe(self) -> Dict[str, Any]:
+        report = {"total": len(self._thousand_eyes), "status": "stable", "eyes": []}
+        for eye in self._thousand_eyes:
+            pulse = eye.gaze()
+            report["eyes"].append(pulse.dict())
+            if pulse.status == "down":
+                report["status"] = "critical"
+        return report
+
+    def _start_dashboard_thread(self, port):
+        def run_server():
+            app = FastAPI(title="Saksh Sidecar")
+            
+            app.add_middleware(
+                CORSMiddleware,
+                allow_origins=["*"],
+                allow_methods=["*"],
+                allow_headers=["*"],
+            )
+
+            @app.get("/api/status")
+            def get_status():
+                return self.observe()
+
+            # --- PATH FIX FOR LIBRARY ---
+            # Finds the 'dashboard/static' folder relative to THIS file (core.py)
+            current_dir = os.path.dirname(os.path.abspath(__file__))
+            static_dir = os.path.join(current_dir, "dashboard", "static")
+            
+            if os.path.exists(static_dir):
+                app.mount("/", StaticFiles(directory=static_dir, html=True), name="static")
+            else:
+                print(f"⚠️ SAKSH WARNING: Static files not found at {static_dir}")
+
+            print(f"\n👁️  SAKSH DASHBOARD ONLINE: http://localhost:{port}\n")
+            
+            config = uvicorn.Config(app, host="0.0.0.0", port=port, log_level="error", loop="asyncio")
+            server = uvicorn.Server(config)
+            server.install_signal_handlers = lambda: None
+            server.run()
+
+        thread = threading.Thread(target=run_server, daemon=True)
+        thread.start()

saksh-1.0.0/saksh/dashboard/server.py

@@ -0,0 +1,48 @@
+from fastapi import FastAPI
+from fastapi.staticfiles import StaticFiles
+from fastapi.middleware.cors import CORSMiddleware
+import uvicorn
+import sys
+import os
+
+# Add root directory to sys.path so we can import 'core' and 'agents'
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+
+from core.saksh import Saksh
+from core.adapters import CrewAINetra
+from agents.qwen_agent import get_qwen_agent
+from agents.gpt_agent import get_gpt_agent
+
+app = FastAPI(title="Saksh Observability")
+
+# Enable CORS (Allows local development flexibility)
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# --- INITIALIZE SAKSH ---
+# We instantiate the observer here. It acts as a standalone monitor.
+observer = Saksh()
+qwen = get_qwen_agent()
+gpt = get_gpt_agent()
+
+# Register Eyes
+observer.open_eye(CrewAINetra(qwen, "Qwen-Custom-01"))
+observer.open_eye(CrewAINetra(gpt, "OpenAI-GPT4-01"))
+
+# --- API ENDPOINTS ---
+
+@app.get("/api/status")
+def get_status():
+    """Returns the JSON health matrix for the UI to consume"""
+    return observer.observe()
+
+# --- SERVE STATIC FILES (HTML/JS/CSS) ---
+app.mount("/", StaticFiles(directory="dashboard/static", html=True), name="static")
+
+if __name__ == "__main__":
+    print("👁️ SAKSH Dashboard running at http://localhost:2604")
+    uvicorn.run(app, host="0.0.0.0", port=2604)

saksh-1.0.0/saksh/dashboard/static/index.html

@@ -0,0 +1,61 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Saksh | Mission Control</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
+    <link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@400;700&display=swap" rel="stylesheet">
+    <link rel="stylesheet" href="style.css">
+</head>
+<body class="bg-black text-light">
+
+    <nav class="navbar navbar-dark border-bottom border-secondary p-3 mb-4" style="background-color: #0d1117;">
+        <div class="container">
+            <a class="navbar-brand fw-bold d-flex align-items-center" href="#">
+                <span class="pulse-dot me-3"></span>
+                👁️ SAKSH <span class="badge bg-secondary ms-2 small-badge">OBSERVER</span>
+            </a>
+            <span class="navbar-text text-success" id="system-status" style="font-size: 0.9rem;">
+                ● System Online
+            </span>
+        </div>
+    </nav>
+
+    <div class="container">
+        
+        <div class="row mb-5 text-center">
+            <div class="col-md-4">
+                <div class="stat-box">
+                    <small class="text-uppercase text-muted fw-bold" style="letter-spacing: 1px;">Active Agents</small>
+                    <h1 id="total-agents" class="display-4 fw-bold text-white mt-2">--</h1>
+                </div>
+            </div>
+            <div class="col-md-4">
+                <div class="stat-box">
+                    <small class="text-uppercase text-muted fw-bold" style="letter-spacing: 1px;">Ecosystem Status</small>
+                    <h1 id="overall-health" class="display-4 fw-bold mt-2">--</h1>
+                </div>
+            </div>
+            <div class="col-md-4">
+                <div class="stat-box">
+                    <small class="text-uppercase text-muted fw-bold" style="letter-spacing: 1px;">Last Heartbeat</small>
+                    <h1 id="last-update" class="display-6 mt-3 text-white">--:--:--</h1>
+                </div>
+            </div>
+        </div>
+
+        <div class="d-flex justify-content-between align-items-center mb-3 border-bottom border-secondary pb-2">
+            <h5 class="text-muted m-0 text-uppercase">Live Agent Telemetry</h5>
+            <span class="badge bg-dark border border-secondary text-muted">AUTO-REFRESH: 2s</span>
+        </div>
+
+        <div class="row g-4" id="agent-grid">
+            </div>
+
+    </div>
+
+    <script src="https://code.jquery.com/jquery-3.6.0.min.js"></script>
+    <script src="script.js"></script>
+</body>
+</html>

saksh-1.0.0/saksh/dashboard/static/script.js

@@ -0,0 +1,68 @@
+$(document).ready(function() {
+    
+    function fetchPulse() {
+        $.ajax({
+            url: '/api/status',
+            method: 'GET',
+            success: function(data) {
+                updateDashboard(data);
+                $('#system-status').html('<span class="text-success">● System Online</span>');
+            },
+            error: function(err) {
+                $('#system-status').html('<span class="text-danger">⚠️ Connection Lost</span>');
+                $('.pulse-dot').css('background-color', 'red');
+            }
+        });
+    }
+
+    function updateDashboard(data) {
+        // 1. Update Header Stats
+        $('#total-agents').text(data.total);
+        
+        let healthColor = data.status === "stable" ? "text-healthy" : "text-down";
+        $('#overall-health').html(`<span class="${healthColor}">${data.status.toUpperCase()}</span>`);
+
+        let now = new Date();
+        $('#last-update').text(now.toLocaleTimeString());
+        
+        // 2. Generate Grid HTML
+        let gridHtml = '';
+        data.eyes.forEach(agent => {
+            let statusColor = agent.status === 'healthy' ? 'text-healthy' : 
+                              agent.status === 'degraded' ? 'text-degraded' : 'text-down';
+            
+            let borderColor = `border-${agent.status}`; // references css class
+            
+            let icon = agent.status === 'healthy' ? '⚡' : '⚠️';
+            if(agent.status === 'down') icon = '🛑';
+
+            gridHtml += `
+                <div class="col-md-4">
+                    <div class="agent-card ${borderColor} p-4 h-100">
+                        <div class="d-flex justify-content-between align-items-start mb-3">
+                            <h5 class="fw-bold mb-0 text-white">${icon} ${agent.agent_id}</h5>
+                            <span class="badge bg-dark border border-secondary">${agent.latency_ms} ms</span>
+                        </div>
+                        
+                        <div class="text-center py-2">
+                            <div class="vitality-number ${statusColor}">${agent.vitality_score}</div>
+                            <small class="text-secondary text-uppercase tracking-wider">Vitality Score</small>
+                        </div>
+
+                        <div class="mt-3 pt-3 border-top border-secondary d-flex justify-content-between">
+                            <small class="text-muted">STATUS</small>
+                            <small class="fw-bold ${statusColor} text-uppercase">${agent.status}</small>
+                        </div>
+                    </div>
+                </div>
+            `;
+        });
+
+        // 3. Update DOM (Only if changed to prevent flicker, simplified here to direct replace)
+        $('#agent-grid').html(gridHtml);
+    }
+
+    // Start Polling
+    fetchPulse();
+    setInterval(fetchPulse, 2000);
+});

saksh-1.0.0/saksh/dashboard/static/style.css

@@ -0,0 +1,77 @@
+/* style.css */
+body {
+    font-family: 'JetBrains Mono', monospace;
+    background-color: #050505 !important;
+    color: #e0e0e0;
+}
+
+/* --- Text Visibility Fixes --- */
+/* Overriding Bootstrap's text-muted to be visible on black */
+.text-muted {
+    color: #8b949e !important; /* Lighter Gray */
+}
+
+h5.text-muted {
+    color: #8b949e !important;
+    font-weight: 600;
+    letter-spacing: 1px;
+}
+/* ----------------------------- */
+
+/* Navbar Pulse Dot */
+.pulse-dot {
+    width: 10px;
+    height: 10px;
+    background-color: #00ff00;
+    border-radius: 50%;
+    box-shadow: 0 0 0 0 rgba(0, 255, 0, 0.7);
+    animation: pulse-green 2s infinite;
+}
+
+@keyframes pulse-green {
+    0% { transform: scale(0.95); box-shadow: 0 0 0 0 rgba(0, 255, 0, 0.7); }
+    70% { transform: scale(1); box-shadow: 0 0 0 10px rgba(0, 255, 0, 0); }
+    100% { transform: scale(0.95); box-shadow: 0 0 0 0 rgba(0, 255, 0, 0); }
+}
+
+/* Stat Boxes */
+.stat-box {
+    background: #0d1117;
+    border: 1px solid #30363d;
+    border-radius: 6px;
+    padding: 20px;
+    height: 100%; /* Ensure equal height */
+}
+
+/* Agent Cards */
+.agent-card {
+    background: #0d1117;
+    border: 1px solid #30363d;
+    border-radius: 8px;
+    transition: all 0.3s ease;
+}
+
+.agent-card:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 4px 15px rgba(0,0,0,0.5);
+}
+
+/* Dynamic Status Borders */
+.border-healthy { border-top: 4px solid #2ea043 !important; }
+.border-degraded { border-top: 4px solid #d29922 !important; }
+.border-down { border-top: 4px solid #da3633 !important; }
+
+/* Text Colors */
+.text-healthy { color: #2ea043; }
+.text-degraded { color: #d29922; }
+.text-down { color: #da3633; }
+
+.vitality-number {
+    font-size: 3.5rem; /* Slightly larger */
+    font-weight: 700;
+}
+
+.small-badge {
+    font-size: 0.6em;
+    vertical-align: middle;
+}

saksh-1.0.0/saksh.egg-info/PKG-INFO

@@ -0,0 +1,118 @@
+Metadata-Version: 2.4
+Name: saksh
+Version: 1.0.0
+Summary: Sahasraksh - The All-Seeing Observer for AI Agents
+Author-email: Ram Bikkina <itsrambikkina@gmail.com>
+Keywords: observability,ai-agents,crewai,langgraph
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: fastapi
+Requires-Dist: uvicorn
+Requires-Dist: requests
+Requires-Dist: pydantic
+Requires-Dist: crewai
+Requires-Dist: build>=1.3.0
+Requires-Dist: twine>=6.0.1
+
+# 👁️ Saksh (Sahasraksh) v1.0.0
+
+> **"The All-Seeing Observer for your AI Agents."**
+
+**Saksh** (derived from *Sahasraksh* — "The One with a Thousand Eyes") is a lightweight, non-intrusive observability layer designed specifically for modern AI Agent ecosystems.
+
+Think of it as a **Sidecar** for your agents. whether you're running a swarm of **CrewAI** workers, a complex **LangGraph** state machine, or raw LLM chains, Saksh sits quietly in the background, monitoring health, latency, and vitality without ever blocking your main event loop.
+
+![Saksh Dashboard](./saksh/dashboard/static/image.png)
+
+
+## ⚡ Why Saksh?
+
+Building agents is hard. Debugging why they "hung" or "timed out" is harder.
+Saksh gives you a **Mission Control** dashboard out of the box, so you can stop guessing if your agents are actually working or just hallucinating silence.
+
+* **Universal & Agnostic:** We don't care if you use CrewAI, LangGraph, or raw OpenAI calls. If it has an API key and a URL, we can watch it.
+* **Zero-Touch Dashboard:** Just instantiate the class, and a full React/FastAPI dashboard spins up on port `2604`. No React code required.
+* **Vitality Scoring:** We don't just check "Is it up?". We calculate a **Vitality Score (0-100)** based on latency spikes and auth failures.
+* **Non-Blocking Sidecar:** Runs in a daemon thread. Your agents keep working even if Saksh takes a coffee break.
+
+## 📦 Installation
+
+Get it via pip:
+
+```bash
+pip install saksh
+
+```
+
+## 🛠️ Quick Start
+
+You can attach Saksh to your existing stack in about 3 lines of code.
+
+### 1. The Setup
+
+```python
+from saksh import Saksh, CrewAINetra, LangGraphNetra
+from crewai import Agent
+
+# 1. Summon the Observer 👁️
+# This auto-magically starts the dashboard at http://localhost:2604
+observer = Saksh(start_dashboard=True)
+
+```
+
+### 2. Connect CrewAI
+
+Saksh automatically detects the LLM configuration (URL, API Key) inside your CrewAI agents.
+
+```python
+# Your standard CrewAI setup
+researcher = Agent(
+    role="Researcher", 
+    goal="Analyze market trends", 
+    backstory="You are a data wizard."
+)
+
+# Open an Eye on it
+observer.open_eye(CrewAINetra(researcher, "Market-Researcher-01"))
+
+```
+
+### 3. Connect LangGraph
+
+Since LangGraph is stateful and abstract, we monitor the underlying LLM provider it relies on.
+
+```python
+# graph_app = workflow.compile()
+
+# Register the graph (defaults to checking OpenAI availability)
+observer.open_eye(LangGraphNetra(graph_app, "Math-Graph-State-Machine"))
+
+```
+
+*That's it. Your terminal will now log:*
+`👁️ Saksh is watching...`
+
+## 🧩 Architecture
+
+Saksh uses a **Netra (Eye) Adapter Pattern**.
+
+* **Saksh (Core):** The singleton observer that manages the background thread and dashboard.
+* **Netra (Adapter):** A standardized interface that knows how to "Gaze" at a specific type of agent and normalize its health data into a `HealthPulse`.
+
+## 🤝 Contributing
+
+This is **v1.0.0** — the "It Works on My Machine" release.
+We are actively looking for contributors to build **Netras** (Adapters) for:
+
+* Autogen
+* Semantic Kernel
+* Local Ollama/Llama.cpp instances
+
+PRs are welcome! Let's build the standard for Agent Observability together.
+
+## 📜 License
+
+MIT © 2026 Ram Bikkina

saksh-1.0.0/saksh.egg-info/SOURCES.txt

@@ -0,0 +1,16 @@
+MANIFEST.in
+README.md
+pyproject.toml
+saksh/__init__.py
+saksh/adapters.py
+saksh/core.py
+saksh.egg-info/PKG-INFO
+saksh.egg-info/SOURCES.txt
+saksh.egg-info/dependency_links.txt
+saksh.egg-info/requires.txt
+saksh.egg-info/top_level.txt
+saksh/dashboard/server.py
+saksh/dashboard/static/image.png
+saksh/dashboard/static/index.html
+saksh/dashboard/static/script.js
+saksh/dashboard/static/style.css

saksh-1.0.0/saksh.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

saksh-1.0.0/saksh.egg-info/requires.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+requests
+pydantic
+crewai
+build>=1.3.0
+twine>=6.0.1

saksh-1.0.0/saksh.egg-info/top_level.txt

@@ -0,0 +1 @@
+saksh

saksh-1.0.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+