@gangtisegts

gangtise-agent wraps a markdown research document api with nine agent types covering company snapshots, investment theses, peer analysis, earnings reviews, and research curation. supports synchronous calls and two-stage async polling for long-running requests.

gangtise-kb-api performs semantic search across internal knowledge bases (reports, opinions, meeting notes) and returns high-relevance content fragments, optimized for extracting document substance rather than file location.

gangtise-file indexes documents (reports, announcements, summaries, opinions, events) in gangtise's file center by type, date, security, institution and keyword. returns file ids with metadata and summaries. differentiates itself from gangtise-kb by focusing on document discovery and filtering rather than content extraction.

gangtise-knowledgebase performs semantic search across internal documents (reports, memos, viewpoints) returning high-relevance content snippets. designed for retrieving document content and key passages rather than file-level filtering or metadata-based discovery.

通过 Gangtise 金融 Open API 拉取结构化量化数据，包括日K行情、财务三大报表、主营构成与估值分位等。需在 scripts 目录配置授权（.authorization），证券参数须为完整代码（如 600519.SH）。当用户需要可落盘的表格化行情与基本面数据时使用。