Overview
[Product intro] Cerul is a video semantic search API that gives AI Agents "vision." It closes the blind spot where traditional AI search cannot understand video content directly and no longer depends on external text glosses. By indexing speech, visual scenes, and on-screen text, it enables deep semantic match. It recognizes slides, charts, and whiteboard sketches in video and turns them into Agent-callable context. The product turns unstructured video into retrievable assets so AI can extract evidence by precise timestamps. [Features] Full-modal semantic retrieval: search across speech, visuals, and on-screen text—even when you cannot recall the exact words. High-precision location feedback: one API call returns structured results with timestamps, relevance scores, and original links. Agent-native integration: MCP protocol support and SDK interfaces for seamless Claude Code, Cursor, and other mainstream Agent frameworks. Industrial retrieval performance: average 150ms low latency for real-time indexing and query on large video libraries.
Connected profiles
Sources · 0Sources & provenance
English text is maintained separately from the original record. Consult the original source before citing a fact.
No source links are available for this profile.
Original source text
Names may use phonetic renderings when no English name is supplied. The original wording is retained below.
【产品介绍】 Cerul 是一款旨在赋予 AI Agent “视觉”能力的视频语义搜索 API。 它解决了传统 AI 搜索无法直接理解视频内容的盲区，使其不再依赖于外部文本解读。 该工具通过索引视频中的语音、视觉场景以及屏幕上的文字信息，实现深层语义匹配。 它能识别视频中出现的幻灯片、图表及白板草图，并将其转化为 Agent 可调用的上下文。 产品致力于将非结构化的视频数据转化为可检索的资产，让 AI 能够基于精准的时间戳提取论据。【产品功能】全模态语义检索：支持跨语音、画面及屏幕文本的综合搜索，即便用户记不清原话也能找回相关片段。高精度定位反馈：单次 API 调用即可返回带时间戳、相关性评分及原始链接的结构化结果。Agent 原生集成：提供 MCP 协议支持与 SDK 接口，可无缝接入 Claude Code、Cursor 等主流 Agent 框架。工业级检索性能：具备平均 150ms 的低延迟表现，支持大规模视频库的实时索引与查询。