docs update for multi llm support + mintlify upgrade

docs.json

@@ -0,0 +1,149 @@
+{
+  "$schema": "https://mintlify.com/docs.json",
+  "theme": "mint",
+  "name": "Potpie",
+  "colors": {
+    "primary": "#0D9373",
+    "light": "#07C983",
+    "dark": "#0D9373"
+  },
+  "favicon": "/favicon.svg",
+  "navigation": {
+    "tabs": [
+      {
+        "tab": "Documentation",
+        "groups": [
+          {
+            "group": "Get Started",
+            "pages": [
+              "introduction",
+              "quickstart"
+            ]
+          },
+          {
+            "group": "How to use Agents",
+            "pages": [
+              "agents/introduction",
+              "agents/debugging-agent",
+              "agents/qna-agent",
+              "agents/integration-test-agent",
+              "agents/unit-test-agent",
+              "agents/code-changes-agent"
+            ]
+          },
+          {
+            "group": "API Access ",
+            "pages": [
+              "agents/api-access"
+            ]
+          },
+          {
+            "group": "Custom Agents",
+            "pages": [
+              "custom-agents/introduction",
+              "custom-agents/configuration",
+              {
+                "group": "Tools",
+                "pages": [
+                  "custom-agents/tools",
+                  "custom-agents/tools/get_code_from_probable_node_name",
+                  "custom-agents/tools/get_code_from_node_id",
+                  "custom-agents/tools/get_code_from_multiple_node_ids",
+                  "custom-agents/tools/ask_knowledge_graph_queries",
+                  "custom-agents/tools/get_nodes_from_tags",
+                  "custom-agents/tools/get_code_from_node_name",
+                  "custom-agents/tools/get_code_graph_from_node_id",
+                  "custom-agents/tools/get_code_graph_from_node_name",
+                  "custom-agents/tools/change_detection"
+                ]
+              }
+            ]
+          }
+        ]
+      },
+      {
+        "tab": "Open Source",
+        "openapi": {
+          "source": "https://production-api.potpie.ai/openapi.json",
+          "directory": "open-source"
+        },
+        "groups": [
+          {
+            "group": "Open Source",
+            "pages": [
+              "open-source/setup",
+              "open-source/getting-started",
+              "open-source/llms",
+              "open-source/ollama-integration/ollama"
+            ]
+          },
+          {
+            "group": "Partners",
+            "pages": [
+              {
+                "group": "AgentOps-AI",
+                "pages": [
+                  "open-source/agentops/agentstack-qna",
+                  "open-source/agentops/agentstack-lld",
+                  "open-source/agentops/agentstack-unittest"
+                ]
+              },
+              {
+                "group": "CrewAIInc",
+                "pages": [
+                  "open-source/crew-ai/crewai-qna",
+                  "open-source/crew-ai/crewai-lld",
+                  "open-source/crew-ai/crewai-unittest"
+                ]
+              }
+            ]
+          }
+        ]
+      }
+    ],
+    "global": {
+      "anchors": [
+        {
+          "anchor": "Documentation",
+          "href": "https://docs.potpie.ai",
+          "icon": "book-open-cover"
+        },
+        {
+          "anchor": "Community",
+          "href": "https://discord.gg/ryk5CMD5v6",
+          "icon": "discord"
+        },
+        {
+          "anchor": "Blog",
+          "href": "https://www.potpie.ai/blog",
+          "icon": "newspaper"
+        }
+      ]
+    }
+  },
+  "logo": {
+    "light": "/logo/light.png",
+    "dark": "/logo/dark.png"
+  },
+  "navbar": {
+    "links": [
+      {
+        "label": "Support",
+        "href": "mailto:[email protected]."
+      }
+    ],
+    "primary": {
+      "type": "button",
+      "label": "Star us on GitHub ⭐️ ",
+      "href": "https://github.com/potpie-ai/potpie"
+    }
+  },
+  "footer": {
+    "socials": {
+      "x": "https://x.com/potpiedotai",
+      "github": "https://github.com/potpie-ai/potpie",
+      "linkedin": "https://www.linkedin.com/company/potpieai",
+      "discord": "https://discord.gg/ryk5CMD5v6"
+    }
+  }
+}

open-source/llms.mdx

@@ -0,0 +1,77 @@
+---
+title: "LLMs"
+description: 'Setup for multiple LLMs'
+---
+
+
+Potpie is designed to work with various large language models (LLMs).This documentation outlines the types of models supported by Potpie, how to configure them.
+
+## Supported Models
+
+Potpie supports a variety of Large Language Models (LLMs). Below is a list of the supported models, categorized by provider and size:
+
+---
+
+### 1. **OpenAI Models**
+
+- **Small Model:**  
+  `openai/gpt-4o-mini`
+
+- **Large Model:**  
+  `openai/gpt-4o`
+
+---
+
+### 2. **Anthropic Models**
+
+- **Small Model:**  
+  `anthropic/claude-3-5-haiku-20241022`
+
+- **Large Model:**  
+  `anthropic/claude-3-7-sonnet-20250219`
+
+---
+
+### 3. **DeepSeek Models**
+
+- **Small Model:**  
+  `openrouter/deepseek/deepseek-chat`
+
+- **Large Model:**  
+  `openrouter/deepseek/deepseek-chat`
+
+---
+
+### 4. **Meta Llama Models**
+
+- **Small Model:**  
+  `openrouter/meta-llama/llama-3.3-70b-instruct`
+
+- **Large Model:**  
+  `openrouter/meta-llama/llama-3.3-70b-instruct`
+
+---
+
+### 5. **Google Gemini Models**
+
+- **Small Model:**  
+  `openrouter/google/gemini-2.0-flash-001`
+
+- **Large Model:**  
+  `openrouter/google/gemini-2.0-flash-001`
+
+---
+
+## Configuration
+
+### Setting Up API Keys
+
+Before using any model, ensure that the appropriate API keys are set up. Potpie checks for API keys in the following order:
+
+1. **Environment Variables:**
+   - `LLM_API_KEY`
+   - `OPENAI_API_KEY`
+   - `{PROVIDER}_API_KEY` (e.g., `ANTHROPIC_API_KEY`)
+
+2. **Secret Manager:**  
+   Potpie can retrieve API keys from a secret management service. Ensure that the keys are stored correctly for the user.

open-source/ollama-integration/ollama.mdx

@@ -0,0 +1,74 @@
+---
+title: "Using Potpie with Ollama Models"
+description: 'Configure and run Potpie with Ollama models for local use.'
+---
+
+# Running Potpie with Ollama Models
+
+In this guide, you'll learn how to configure and run Potpie with Ollama models on your local machine.
+
+## Step 1: Install Ollama
+
+Before using Ollama models with Potpie, you need to install the **Ollama CLI** tool. Ollama allows you to run models locally, and you can install it with the following commands:
+
+### Installation:
+
+1. Open a terminal and run the following command to download and install Ollama:
+   ```bash
+   curl -fsSL https://ollama.com/install.sh | sh
+   ```
+2. Once installed, verify the installation by running:
+  ```bash
+   ollama --version
+   ```
+
+## Step 2: Set Up Ollama Models
+
+### Step 2.1: Pull the Required Models
+
+To run Potpie with Ollama, you need to download the models you plan to use. In this guide, we pull two models commonly used for low and high reasoning tasks:
+
+- **Low Reasoning Model:** Used for generating the knowledge graph.
+- **High Reasoning Model:** Used for agent reasoning.
+
+Run the following commands to pull the models:
+
+```bash
+# Pull the low reasoning model (used for knowledge graph generation)
+ollama pull ollama_chat/qwen2.5-coder:7b
+
+# Pull the high reasoning model (used for agent reasoning)
+ollama pull ollama_chat/qwen2.5-coder:7b
+```
+
+Note that the models you pull should be in the `provider/model_name` format, or they should be compatible with the format expected by Litellm. For more model options and details, refer to the [Litellm documentation](https://docs.litellm.ai/).
+
+## Step 3: Ollama API Key
+
+You can retrieve your API Key following these steps:
+
+1. **Sign in to Ollama**: Go to Ollama's official website and sign in to your account.
+2. **Find Your API Key**: Navigate to the [ollama keys](https://ollama.com/settings/keys).
+3. **Find the API Key Path**: You can now copy your API Key.
+
+
+## Step 4: Configure Environment Variables
+
+Once you have installed Ollama and pulled the models, you need to configure your environment to use these models with Potpie.
+
+Open or create a `.env` file in the directory where you're running Potpie. Add the following configuration to specify the Ollama models:
+
+```bash
+# Set the LLM provider to Ollama
+LLM_PROVIDER=ollama
+
+# Set the API key 
+LLM_API_KEY=PASTE-YOUR-API-KEY-HERE
+
+# Specify the model used for low reasoning (knowledge graph generation)
+LOW_REASONING_MODEL=ollama_chat/qwen2.5-coder:7b
+
+# Specify the model used for high reasoning (agent reasoning)
+HIGH_REASONING_MODEL=ollama_chat/qwen2.5-coder:7b
+```
+All Set! Potpie will now use Local Ollama Models.