Releases · potpie-ai/potpie

What's Changed

We are excited to announce the first official release of Potpie, a Prompt-To-Agent platform designed to understand your codebase, and provide the tooling required by the agents to accomplish any engineering task.

Key Features:

Comprehensive Codebase Analysis: Potpie constructs a detailed knowledge graph of your codebase, capturing relationships between functions, classes, and files to facilitate intelligent code analysis and manipulation.
Pre-built AI Agents: Utilize specialized agents to automate common engineering tasks:
- Debugging Agent: Analyzes stack traces and provides tailored debugging steps.
- Codebase Q&A Agent: Answers questions about your codebase, explaining functions, features, setup and architecture.
- Code Changes Agent: Evaluates code changes, identifies blast radius of changes, and suggests improvements.
- Integration Test Agent: Generates integration test plans and code to ensure component compatibility.
- Unit Test Agent: Creates unit test plans and code to enhance test coverage.
- Low-Level Design (LLD) Agent: Develops low-level designs for new features based on functional requirements.
Custom Agent Development Platform: Beyond pre-built agents, Potpie offers a platform to create custom agents tailored to your project's unique needs, enabling automation of specific, repeatable tasks.
Extensible Tooling System: Potpie provides a set of tools that agents can use to interact with the knowledge graph and the underlying infrastructure. These tools are vital for creating custom agents and for performing highly contextual tasks with precision. The available tools include:
- get_code_from_probable_node_name: Retrieves code snippets based on a probable node name from the knowledge graph.
- get_code_from_node_id: Fetches code snippets associated with a specific node ID.
- get_code_from_multiple_node_ids: Allows retrieval of code snippets for multiple node IDs simultaneously.
- ask_knowledge_graph_queries: Executes vector similarity search queries against the knowledge graph to obtain relevant information.
- get_nodes_from_tags: Retrieves nodes that are associated with specific tags in the knowledge graph.
- get_code_from_node_name: Obtains code snippets linked to a specific node name.
- get_code_graph_from_node_id: Fetches the code graph structure related to a particular node ID (works best with Python, JavaScript, TypeScript).
- get_code_graph_from_node_name: Retrieves the code graph structure associated with a specific node name (works best with Python, JavaScript, TypeScript).
- change_detection: Detects changes in the current branch with respect to the default branch of the repository and retrieves updated function details.