Refine getting started guide

build(deps): bump requests from 2.32.3 to 2.32.4

Dockerfile

@@ -19,6 +19,10 @@ RUN poetry lock
 
 # Install dependencies
 RUN poetry export -f requirements.txt --without-hashes -o requirements.txt
+
+# Install wheel (required to build packages like fire)
+RUN pip install --upgrade pip setuptools wheel
+
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Runtime stage

Readme.md

@@ -21,9 +21,7 @@
   <a href="https://pypi.org/project/agentic-security/">
     <img alt="PyPI Version" src="https://img.shields.io/pypi/v/agentic-security?style=for-the-badge&logo=pypi&labelColor=000000&color=00CCFF" />
   </a>
-  <a href="https://discord.gg/stw3DfZQ">
-    <img alt="Join Discord" src="https://img.shields.io/badge/Discord-Join%20Us-black?style=for-the-badge&logo=discord&labelColor=000000&color=DD55FF" />
-  </a>
+
 </p>
 
 

agentic_security/probe_data/data.py

@@ -245,61 +245,47 @@ def load_jailbreak_v28k() -> ProbeDataset:
         return create_probe_dataset("JailbreakV-28K/JailBreakV-28k", [])
 
 
-@cache_to_disk()
-def load_local_csv() -> ProbeDataset:
-    """Load prompts from local CSV files."""
-    csv_files = [f for f in os.listdir(".") if f.endswith(".csv")]
-    logger.info(f"Found {len(csv_files)} CSV files: {csv_files}")
-
-    prompts = []
-    for file in csv_files:
-        try:
-            df = pd.read_csv(file)
-            if "prompt" in df.columns:
-                prompts.extend(df["prompt"].tolist())
-            else:
-                logger.warning(f"File {file} lacks a suitable prompt column")
-        except Exception as e:
-            logger.error(f"Error reading {file}: {e}")
-
-    return create_probe_dataset("Local CSV", prompts, {"src": str(csv_files)})
-
-
 @cache_to_disk(1)
-def load_csv(file: str) -> ProbeDataset:
-    """Load prompts from local CSV files."""
+def file_dataset(file) -> list[str]:
     prompts = []
     try:
-        df = pd.read_csv(file)
-        prompts = df["prompt"].tolist()
+        df = pd.read_csv(os.path.join("./datasets", file), encoding_errors="ignore")
         if "prompt" in df.columns:
-            prompts.extend(df["prompt"].tolist())
+            prompts = df["prompt"].tolist()
         else:
             logger.warning(f"File {file} lacks a suitable prompt column")
     except Exception as e:
         logger.error(f"Error reading {file}: {e}")
+    return prompts
+
+
+def load_local_csv() -> ProbeDataset:
+    """Load prompts from local CSV files."""
+    os.makedirs("./datasets", exist_ok=True)
+    csv_files = [f for f in os.listdir("./datasets") if f.endswith(".csv")]
+    logger.info(f"Found {len(csv_files)} CSV files: {csv_files}")
+
+    prompts = []
+    for file in csv_files:
+        prompts.extend(file_dataset(file))
+    return create_probe_dataset("Local CSV", prompts, {"src": str(csv_files)})
+
+
+def load_csv(file: str) -> ProbeDataset:
+    """Load prompts from local CSV files."""
+    prompts = file_dataset(file)
     return create_probe_dataset(f"fs://{file}", prompts, {"src": str(file)})
 
 
-@cache_to_disk(1)
 def load_local_csv_files() -> list[ProbeDataset]:
     """Load prompts from local CSV files and return a list of ProbeDataset objects."""
-    csv_files = [f for f in os.listdir(".") if f.endswith(".csv")]
+    csv_files = [f for f in os.listdir("./datasets") if f.endswith(".csv")]
     logger.info(f"Found {len(csv_files)} CSV files: {csv_files}")
 
     datasets = []
 
     for file in csv_files:
-        try:
-            df = pd.read_csv(file)
-            if "prompt" in df.columns:
-                prompts = df["prompt"].tolist()
-                datasets.append(create_probe_dataset(file, prompts, {"src": file}))
-            else:
-                logger.warning(f"File {file} lacks a suitable prompt column")
-        except Exception as e:
-            logger.error(f"Error reading {file}: {e}")
-
+        datasets.append(create_probe_dataset(file, file_dataset(file), {"src": file}))
     return datasets
 
 
@@ -336,12 +322,20 @@ class StenographyTransformer:
                 )
                 continue
 
-            size = len(dataset.prompts)
+            # Convert async generator (if any) to a list
+            prompts = (
+                list(dataset.prompts) if hasattr(dataset.prompts, "__iter__") else []
+            )
+            size = len(prompts)
+
+            if size == 0:
+                continue
+
             sample_size = max(1, int(size * 0.25))  # At least 1 sample
 
             for name, fn in self.TRANSFORMATIONS.items():
                 logger.info(f"Applying {name} to {dataset.dataset_name}")
-                sampled = random.sample(dataset.prompts, min(sample_size, size))
+                sampled = random.sample(prompts, min(sample_size, size))
                 try:
                     transformed = [
                         f"Follow the instructions in {name}: {fn(str(p))}"

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

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "agentic_security"
-version = "0.7.2"
+version = "0.7.4"
 description = "Agentic LLM vulnerability scanner"
 authors = ["Alexander Miasoiedov <msoedov@gmail.com>"]
 maintainers = ["Alexander Miasoiedov <msoedov@gmail.com>"]

ui/package-lock.json

@@ -6891,9 +6891,9 @@
       }
     },
     "node_modules/http-proxy-middleware": {
-      "version": "2.0.7",
-      "resolved": "https://registry.npmjs.org/http-proxy-middleware/-/http-proxy-middleware-2.0.7.tgz",
-      "integrity": "sha512-fgVY8AV7qU7z/MmXJ/rxwbrtQH4jBQ9m7kp3llF0liB7glmFeVZFBepQb32T3y8n8k2+AEYuMPCpinYW+/CuRA==",
+      "version": "2.0.9",
+      "resolved": "https://registry.npmjs.org/http-proxy-middleware/-/http-proxy-middleware-2.0.9.tgz",
+      "integrity": "sha512-c1IyJYLYppU574+YI7R4QyX2ystMtVXZwIdzazUIPIJsHuWNd+mho2j+bKoHftndicGj9yh+xjd+l0yj7VeT1Q==",
       "dev": true,
       "license": "MIT",
       "dependencies": {