AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
Sealant is the open-source, self-hosted runtime under your AI agent. One call spins up a real workspace around your repo, runs your harness, and hands back a structured run you can replay. Bring your own agent. Keep your code. Read the evidence yourself.
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
Sealant is an open-source, self-hosted runtime for agentic development. It gives coding
harnesses a real workspace to work in, then turns every run into a structured, replayable
execution record — the change, the checks, the terminal output, the artifacts, the browser
evidence, and the source trail behind the result.
Bring your own harness. Keep your code. Read the evidence yourself.查看完整 README ↗目前无法加载完整 README。请通过原文链接阅读。 以下为已保存的摘录。
Sealant is an open-source, self-hosted runtime for agentic development. It gives coding
harnesses a real workspace to work in, then turns every run into a structured, replayable
execution record — the change, the checks, the terminal output, the artifacts, the browser
evidence, and the source trail behind the result.
Bring your own harness. Keep your code. Read the evidence yourself.查看完整 README ↗