SparkBeyond/agentune
Tune your AI Agent to best meet its KPI with a cyclic process of analyze, improve and simulate
Extracts semantic and structural features from real conversations to identify statistically significant KPI drivers, then generates ranked improvement recommendations for prompts and policies. Includes a separate agentune-simulate library for safe pre-deployment testing of changes. Built for production agents with measurable outcomes, targeting ML/data teams accountable for business metrics rather than model accuracy alone.
Available on PyPI.
Stars
36
Forks
3
Language
Python
License
Apache-2.0
Category
Last pushed
Jan 14, 2026
Monthly downloads
36
Commits (30d)
0
Dependencies
16
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/SparkBeyond/agentune"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Featured in
Related agents
StonyBrookNLP/appworld
🌍 AppWorld: A Controllable World of Apps and People for Benchmarking Function Calling and...
qualifire-dev/rogue
AI Agent Evaluator & Red Team Platform
future-agi/ai-evaluation
Evaluation Framework for all your AI related Workflows
microsoft/WindowsAgentArena
Windows Agent Arena (WAA) 🪟 is a scalable OS platform for testing and benchmarking of...
agentscope-ai/OpenJudge
OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards