Updated workflow image

Author: prane-eth

.dockerignore

@@ -6,22 +6,8 @@ EMBED_MODEL_NAME=sentence-transformers/all-MiniLM-L12-v2
 EMBEDDING_BASE_URL=http://localhost:1234/v1
 EMBEDDING_API_KEY=your-embedding-key
 
-# Backend addresses
-BACKEND_BASE_URL=http://api_server:8000
-BACKEND_API_KEY=change-me
-
 # Use an OpenAI-compatible API
 OPENAI_BASE_URL=https://api.tokenfactory.nebius.com/v1
 OPENAI_API_KEY=<your_api_key_here>
 OPENAI_MODEL=openai/gpt-oss-20b
 OPENAI_MODEL_GUARDED=openai/gpt-oss-20b-GUARDED
-
-MCP_URL_DIRECT=https://<your_mcp_server_url_here>.app/unguarded-mcp/
-MCP_URL_GUARDED=https://<your_mcp_server_url_here>.app/mcp/
-
-EMBED_MODEL_NAME=<your_embedding_model_name_here>
-EMBEDDING_BASE_URL=<your_embedding_base_url_here>
-
-# Optional: ngrok authtoken for the ngrok service (leave empty to skip public tunnel)
-NGROK_AUTHTOKEN=
-NGROK_DOMAIN=

.env.example

@@ -43,8 +43,8 @@ $ ffmpeg -i unused/banner_video.mp4 -vframes 1 project_banner.jpg
 - 💭 Model being overconfident in its incorrect knowledge.
 - 🚧 Lack of proper constraints or guidelines for the agent.
 - 📉 Inadequate training data for specific scenarios.
-- 🛠️ MCP server providing incorrect tool descriptions that mislead the agent.
-- 🎭 Harmful MCP servers returning manipulative text to mislead the model.
+- 🛠️ Tools with incorrect descriptions that mislead the agent.
+- 🎭 Harmful tools descriptions including manipulative text to mislead the model.
 - 😬 The experiments proved that the model performs a harmful action and still responds "Sorry, I can't help with that."
 
 ## 🆕 New contributions of Agent-Action-Guard framework:
@@ -76,13 +76,13 @@ $ ffmpeg -i unused/banner_video.mp4 -vframes 1 project_banner.jpg
 ## ✨ Special features:
 - This project introduces "HarmActionsBench" dataset and benchmark to evaluate an AI agent's probability of generating harmful actions.
 - The dataset has been used to train a lightweight neural network model that classifies actions as safe, harmful, or unethical.
-- ⚡ The model is lightweight and can be easily integrated into existing AI agent frameworks like MCP.
-- 🔌 Supports MCP (Model Context Protocol) to allow real-time action classification.
+- ⚡ The model is lightweight and can be easily integrated into existing AI agent frameworks.
+<!-- - 🔌 Supports MCP (Model Context Protocol) to allow real-time action classification. -->
 <!-- - Unlike OpenAI's `"require_approval": "always"` flag, this blocks harmful actions without human intervention. -->
 <!-- - 🤝 A2A-compatible version: https://github.com/Pro-GenAI/A2A-Agent-Action-Guard. -->
 
 🛡️ **Safety Features:**
-- 🔍 Automatically classifies MCP tool calls before execution.
+- 🔍 Automatically classifies tool calls before execution.
 - 🚫 Blocks harmful actions based on the outputs of the trained model.
 - 📋 Provides detailed classification results.
 - ✅ Allows safe actions to proceed normally.