Multi-step prompt sequences for complex AI workflows.
99 tools found
A notebook for triggering a saved ChatGPT Workspace Agent from an external API, verifying the async run in its destination.
A fast-executing fixture with two local paper records demonstrating legacy tool-calling patterns, derived from the Cookbook arXiv retrieval example.
A lightweight Qdrant vector search fixture with embeddings and a small article set, derived from the Cookbook Qdrant search example.
Promptfoo example evaluating Microsoft's MAI-Image-2.5 model on Azure AI Foundry, with vision-based grading of generated images.
A promptfoo example that scores two Codex skill versions against the same review tasks and auto-selects the stronger output.
Promptfoo workflow that benchmarks gpt-5.4, gpt-5.4-mini, and gpt-5.4-nano side-by-side on identical prompts to compare quality, latency, and cost across
Promptfoo example that compares OpenAI gpt-5.4-mini with two reasoning effort levels (none vs. medium) to evaluate output quality, latency, and cost.
Promptfoo wrapper running the OSWorld computer-use benchmark via Inspect, with reference GPT-5.5 pass-rate data.
A promptfoo example using MLflow AI Gateway as an LLM provider, including as its own grading model.
Promptfoo example for OrcaRouter, an OpenAI-compatible adaptive routing gateway to multiple upstream models.
Promptfoo internal suite testing its own PolicyPlugin redteam test generator across five generation modes.
AI code review CLI from Alibaba that reviews diffs or whole files with line-level precision, using about 1/9 the tokens of general agents.
