promptfoo

Superagent is an open source AI safety and agent development platform that helps developers and organizations build, deploy, and protect AI-driven applications and assistants by embedding safety guardrails, runtime security, and compliance controls into agent workflows. It provides purpose-trained models and APIs (such as Guard, Verify, and Redact) that block prompt injections, malicious tool calls, data leakage, and unsafe outputs in real time, while red-teaming tests probe production systems for vulnerabilities and deliver findings with remediation guidance. Superagent integrates with existing AI systems at inference and tool-call layers to filter inputs/outputs, remove sensitive data like PII/PHI, enforce policy constraints, and stop unauthorized actions before they occur, offering unified observability, live trace logs, policy controls, and audit trails for security and engineering teams.

ChainForge is an open-source visual programming environment designed for prompt engineering and large language model evaluation. It enables users to assess the robustness of prompts and text-generation models beyond anecdotal evidence. Simultaneously test prompt ideas and variations across multiple LLMs to identify the most effective combinations. Evaluate response quality across different prompts, models, and settings to select the optimal configuration for specific use cases. Set up evaluation metrics and visualize results across prompts, parameters, models, and settings, facilitating data-driven decision-making. Manage multiple conversations simultaneously, template follow-up messages, and inspect outputs at each turn to refine interactions. ChainForge supports various model providers, including OpenAI, HuggingFace, Anthropic, Google PaLM2, Azure OpenAI endpoints, and locally hosted models like Alpaca and Llama. Users can adjust model settings and utilize visualization nodes.