PySpur is an AI agent builder written based on Python. AI engineers use it to build agents, implement it step by step and check past operating records. Test-driven: Build a workflow, run test cases, and iterate. 👤People in the loop: Persistence the workflow and wait for manual approval or rejection. Loop: An iterative tool call with memory function. File upload: Upload a file or paste a URL to process the document. Structured output: JSON Schema UI editor. ˇRAG: Parses, blocks, embeds and updates data to a vector database. ˇMultimodal: supports video, images, audio, text, and code. Tools: Slack, Firecrawl.dev, Google Sheets, GitHub, etc. Evaluation: Evaluate agents on real data sets. 🚀One-click deployment: Publish as an API and integrate anywhere. 🐍Python-based: Add new nodes by creating a single Python file. ˇVendor support: Supports more than 100 LLM vendors, embedders and vector databases.
