Vibe Coding DiscoverJIT-Agent
JIT-Agent is a meta-agent that generates and revises executable, task-specific harnesses for agentic LLMs. It includes harness designs, benchmark adapters, and candidate selection for evaluating workflows across research, office, planning, and other tasks.
Use Cases
Generate task-specific agent harnessesAdapt agent workflows using execution feedbackEvaluate agent harnesses across benchmarksBuild agents for deep researchAutomate office document tasksRun planning and workspace agents
Built With
- Language
- Python
- Frameworks
- OpenAI-compatible API · vLLM · SGLang
Tags
meta-agent · agent harnesses · harness generation · harness evolution · task-specific agents · agent evaluation · multi-benchmark