Refine getting started guide

agentic_security/mcp/client.py

@@ -3,8 +3,6 @@ import asyncio
 from mcp import ClientSession, StdioServerParameters
 from mcp.client.stdio import stdio_client
 
-from agentic_security.logutils import logger
-
 # Create server parameters for stdio connection
 server_params = StdioServerParameters(
     command="python",  # Executable
@@ -14,54 +12,42 @@ server_params = StdioServerParameters(
 
 
 async def run() -> None:
-    try:
-        logger.info(
-            "Starting stdio client session with server parameters: %s", server_params
-        )
-        async with stdio_client(server_params) as (read, write):
-            async with ClientSession(read, write) as session:
-                # Initialize the connection --> connection does not work
-                logger.info("Initializing client session...")
-                await session.initialize()
+    async with stdio_client(server_params) as (read, write):
+        async with ClientSession(read, write) as session:
+            # Initialize the connection --> connection does not work
+            await session.initialize()
 
-                # List available prompts, resources, and tools --> no avalialbe tools
-                logger.info("Listing available prompts...")
-                prompts = await session.list_prompts()
-                logger.info(f"Available prompts: {prompts}")
+            # List available prompts, resources, and tools --> no avalialbe tools
+            prompts = await session.list_prompts()
+            print(f"Available prompts: {prompts}")
 
-                logger.info("Listing available resources...")
-                resources = await session.list_resources()
-                logger.info(f"Available resources: {resources}")
+            resources = await session.list_resources()
+            print(f"Available resources: {resources}")
 
-                logger.info("Listing available tools...")
-                tools = await session.list_tools()
-                logger.info(f"Available tools: {tools}")
+            tools = await session.list_tools()
+            print(f"Available tools: {tools}")
 
-                # Call the echo tool --> echo tool issue
-                logger.info("Calling echo_tool with message...")
-                echo_result = await session.call_tool(
-                    "echo_tool", arguments={"message": "Hello from client!"}
-                )
-                logger.info(f"Tool result: {echo_result}")
+            # Call the echo tool --> echo tool iisue
+            echo_result = await session.call_tool(
+                "echo_tool", arguments={"message": "Hello from client!"}
+            )
+            print(f"Tool result: {echo_result}")
 
-                # # Read the echo resource
-                # echo_content, mime_type = await session.read_resource(
-                #     "echo://Hello_resource"
-                # )
-                # logger.info(f"Resource content: {echo_content}")
-                # logger.info(f"Resource MIME type: {mime_type}")
+            # # Read the echo resource
+            # echo_content, mime_type = await session.read_resource(
+            #     "echo://Hello_resource"
+            # )
+            # print(f"Resource content: {echo_content}")
+            # print(f"Resource MIME type: {mime_type}")
 
-                # # Get and use the echo prompt
-                # prompt_result = await session.get_prompt(
-                #     "echo_prompt", arguments={"message": "Hello prompt!"}
-                # )
-                # logger.info(f"Prompt result: {prompt_result}")
+            # # Get and use the echo prompt
+            # prompt_result = await session.get_prompt(
+            #     "echo_prompt", arguments={"message": "Hello prompt!"}
+            # )
+            # print(f"Prompt result: {prompt_result}")
 
-                logger.info("Client operations completed successfully.")
-                return prompts, resources, tools
-    except Exception as e:
-        logger.error(f"An error occurred during client operations: {e}", exc_info=True)
-        raise
+            # You can perform additional operations here as needed
+            return prompts, resources, tools
 
 
 if __name__ == "__main__":

agentic_security/mcp/main.py

@@ -4,6 +4,7 @@ from mcp.server.fastmcp import FastMCP
 # Initialize MCP server
 mcp = FastMCP(
     name="Agentic Security MCP Server",
+    description="MCP server to interact with LLM scanning test",
     dependencies=["httpx"],
 )
 

agentic_security/probe_data/data.py

@@ -8,6 +8,7 @@ from typing import Any, TypeVar
 import httpx
 import pandas as pd
 from cache_to_disk import cache_to_disk
+from datasets import load_dataset
 
 from agentic_security.logutils import logger
 from agentic_security.probe_data import stenography_fn
@@ -19,7 +20,6 @@ from agentic_security.probe_data.modules import (
     inspect_ai_tool,
     rl_model,
 )
-from datasets import load_dataset
 
 # Type aliases for clarity
 T = TypeVar("T")

docs/getting_started.md

@@ -1,23 +1,54 @@
 # Getting Started
 
-Welcome to Agentic Security! This guide will help you get started with using the tool.
+Welcome to Agentic Security! This guide will help you orient yourself within the project and start using the tool quickly.
+
+## Project Overview
+
+Agentic Security is an open-source vulnerability scanner for Large Language Models (LLMs). It provides both a command line interface and a FastAPI server so you can probe models for weaknesses such as jailbreaks or refusal patterns. The tool supports multimodal attacks, multi-step scans and reinforcement-learning based probes.
+
+## Repository Layout
+
+```
+agentic_security/
+├── __main__.py    - CLI entry point
+├── app.py         - FastAPI app assembly
+├── lib.py         - SecurityScanner and utilities
+├── config.py      - Configuration handling
+├── core/          - app state and logging helpers
+├── probe_actor/   - scanning logic and RL modules
+├── probe_data/    - dataset registry and loaders
+├── routes/        - API endpoints
+└── ui/            - Web UI assets (Vue)
+```
+
+`tests/` contains unit tests, and `docs/` houses the project documentation.
 
 ## Quick Start
 
 1. Ensure you have completed the [installation](installation.md) steps.
-1. Run the following command to start the application:
+2. Run the following command to start the application:
    ```bash
    agentic_security
    ```
-1. Access the application at `http://localhost:8718`.
+   The server will start on `http://localhost:8718`.
+3. Explore available commands with:
+   ```bash
+   agentic_security --help
+   ```
 
 ## Basic Usage
 
-- To view available commands, use:
+- To view available commands, run:
   ```bash
   agentic_security --help
   ```
 
 ## Next Steps
 
-Explore the [Configuration](configuration.md) section to customize your setup.
+- Review the [Quickstart Guide](quickstart.md) for a fast setup walkthrough.
+- Check [http_spec.md](http_spec.md) to learn how LLM endpoints are described.
+- Browse the `probe_actor` and `probe_data` modules to understand how scanning works and how datasets are loaded.
+- Explore the [Configuration](configuration.md) section to customize your setup.
+- Run the tests in `tests/` to verify your environment once dependencies are installed.
+
+This guide should give you a solid foundation for exploring and extending Agentic Security. For more details, see the rest of the documentation.

pyproject.toml

@@ -28,50 +28,54 @@ agentic_security = "agentic_security.__main__:main"
 
 [tool.poetry.dependencies]
 python = "^3.11"
-fastapi = "^0.116.1"
-uvicorn = "^0.35.0"
+fastapi = "^0.115.8"
+uvicorn = "^0.34.0"
 fire = "0.7.0"
 loguru = "^0.7.3"
 httpx = "^0.28.1"
 cache-to-disk = "^2.0.0"
 pandas = ">=1.4,<3.0"
-datasets = "^4.0.0"
+datasets = "^3.3.0"
 tabulate = ">=0.8.9,<0.10.0"
 colorama = "^0.4.4"
-matplotlib = "^3.10.5"
-pydantic = "^2.11.7"
+matplotlib = "^3.9.2"
+pydantic = "2.10.6"
 scikit-optimize = "^0.10.2"
-scikit-learn = "^1.7.1"
+scikit-learn = "1.6.1"
 numpy = ">=1.24.3,<3.0.0"
 jinja2 = "^3.1.4"
 python-multipart = "^0.0.20"
 tomli = "^2.2.1"
-rich = "^14.1.0"
+rich = "13.9.4"
 gTTS = "^2.5.4"
-sentry_sdk = "^2.34.1"
-orjson = "^3.11.2"
-pyfiglet = "^1.0.3"
-termcolor = "^3.1.0"
-mcp = "^1.12.4"
+sentry_sdk = "^2.22.0"
+orjson = "^3.10"
+pyfiglet = "^1.0.2"
+termcolor = "^2.4.0"
+mcp = "^1.4.1"
+
 # garak = { version = "*", optional = true }
-pytest-xdist = "^3.8.0"
+pytest-xdist = "3.6.1"
+
 
 [tool.poetry.group.dev.dependencies]
 # Pytest
-pytest = "^8.4.1"
-pytest-asyncio = "^1.1.0"
-inline-snapshot = ">=0.13.3,<0.27.2"
+pytest = "^8.3.4"
+pytest-asyncio = "^0.25.2"
+inline-snapshot = ">=0.13.3,<0.21.0"
 pytest-httpx = "^0.35.0"
-pytest-mock = "^3.14.1"
+pytest-mock = "^3.14.0"
+
 # Rest
 black = ">=24.10,<26.0"
-mypy = "^1.17.1"
-pre-commit = "^4.3.0"
-huggingface-hub = ">=0.25.1,<0.34.4"
+mypy = "^1.12.0"
+pre-commit = "^4.0.1"
+huggingface-hub = ">=0.25.1,<0.30.0"
+
 # Docs
 mkdocs = ">=1.4.2"
-mkdocs-material = "^9.6.16"
-mkdocstrings = ">=0.29.0"
+mkdocs-material = "^9.6.4"
+mkdocstrings = ">=0.26.1"
 mkdocs-jupyter = ">=0.25.1"
 
 

tests/test_registry.py

@@ -1,7 +1,7 @@
 import pytest
+from datasets import load_dataset
 
 from agentic_security.probe_data import REGISTRY
-from datasets import load_dataset
 
 
 @pytest.mark.slow

ui/package.json

@@ -19,7 +19,7 @@
     "@vue/cli-plugin-babel": "~5.0.0",
     "@vue/cli-plugin-eslint": "~5.0.0",
     "@vue/cli-service": "~5.0.0",
-    "eslint": "^9.39.1",
+    "eslint": "^7.32.0",
     "eslint-plugin-vue": "^8.0.3"
   },
   "eslintConfig": {