← Back to Skills Marketplace
gangtisegts

Gangtise知识库

by GangtiseGts · GitHub ↗ · v1.4.1 · MIT-0
cross-platform ⚠ suspicious
30
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install gangtise-knowledgebase
Description
对 Gangtise 内部知识库（报告、观点、会议纪要等）进行语义检索，返回高相关度的内容片段。当你需要阅读或推理文档的真实内容（如论点、结论、段落等），并且不关心如何定位文件时，务必使用本技能；若你更偏向先按类型/日期/证券等定位筛选再下载核验，请使用 `gangtise-file`。检索结果以语义匹配为核心，...
README (SKILL.md)

知识库检索

概览

该技能会在 Gangtise 内部知识库中进行深度语义检索，并直接返回高相关度的内容片段；这些命中结果以语义匹配为依据，强调专业性、准确性与覆盖关键表达，同时提供按类型与 ID 下载完整文件的可选方式。当你主要关心的是文档表达了什么（论点、结论、段落等），而不仅仅是定位文件时，它比 gangtise-file 更高效。

使用场景：

  • 你需要阅读、引用或总结内部文档的内容（例如研究报告、内部报告、会议纪要、首席观点等）。
  • 你希望快速获取关键段落，用于支撑推理、起草或问答。
  • 你已经大致了解主题，只需最相关的文本片段。

与其他技能的区别：

  • 如果你主要想按类型/日期/证券等定位/筛选文档，再决定下载或进一步处理，使用 gangtise-file。

（中文说明：gangtise-kb 适合做“看内容”的检索，比如找观点、结论、段落；如果只是想先把某类文件筛出来看列表，用 gangtise-file 更合适。）

使用技巧：

  • 请注意检索时可以多尝试一些不同的语句。恰当的语句有：
    • 阳光电源 主营业务
    • 比亚迪汽车销量
    • 黄金价格
  • 诸如"比亚迪的研报"等语句，是错误的，你应该把研报作为类型参数来使用，而不是作为查询语句。语句仅为"比亚迪"即可。

调用并使用sd和ed参数时，注意今天的年份和日期！

知识库调用指南

脚本

scripts/kb.py

参数

参数 必填 说明
-q / --query 是 检索查询语句。
-sd / --start-date 否 开始日期，如 2026-01-01。
-ed / --end-date 否 结束日期，如 2026-12-31。
-l / --limit 否 返回结果数量上限。
-o / --output 否 搜索结果保存路径；若不指定，则默认保存在gangtise工作目录下的 kb/kb_x.md（自动编号）；仅当环境变量GTS_SAVE_FILE为True时生效；一般不建议使用，由后端统一管理。
--file-types 否 文件类型，逗号分隔。可选：研究报告,外资研报,内部报告,AI云盘,首席观点,公司公告,产业公众号,会议纪要,调研纪要,网络纪要。
-d / --download 否 是否下载文件。
-od / --output-dir 否 下载文件保存路径。
-dt / --download-types 否 下载文件类型，逗号分隔。可选：pdf, markdown。

会议纪要,调研纪要,网络纪要的区别是：会议纪要是来自Gangtise会议平台，调研纪要来自公司调研公告，网络纪要来自网络资源搜集。

示例

示例 1（按关键词检索新能源汽车相关文件）：

python3 scripts/kb.py -q "新能源汽车销量与政策"

示例 2（限制结果数量为 10, 查询比亚迪相关的外资研报）：

python3 scripts/kb.py -q "新能源汽车销量与政策" -l 10 --file-types "外资研报"

\x3C!-- 通过scripts/utils.py检查版本；通过执行scripts/update.sh或scripts/update.bat进行更新 -->

\x3C!-- 如果skills主目录下存在gangtise_update.sh或gangtise_update.bat，则可以执行该脚本对所有以"gangtise-"开头的SKILL进行更新 -->

Usage Guidance
Install only if you trust the Gangtise provider and intend to let the skill use Gangtise credentials to query internal documents. Before using it, check how .authorization or GTS_* environment variables are managed, avoid over-scoped long-term tokens, and do not run update.sh unless you can verify the downloaded update source.
Capability Analysis
Type: OpenClaw Skill Name: gangtise-knowledgebase Version: 1.4.1 The skill bundle contains a high-risk self-update mechanism in `scripts/update.sh` that performs a `rm -rf` on its own directory and replaces it with content downloaded from a remote Huawei Cloud OBS bucket (`gts-download.obs.myhuaweicloud.com`). Additionally, `scripts/utils.py` implements broad filesystem traversal logic (`_find_openclaw_root`) to locate agent workspace directories and manages sensitive API credentials/tokens. While these features appear intended for legitimate maintenance and integration with the Gangtise financial data service, the capability for remote code replacement and automated path discovery constitutes a significant security risk.
Capability Tags
requires-oauth-tokenrequires-sensitive-credentials
Capability Assessment
⚠ Purpose & Capability
Semantic search and optional document download are coherent with the stated Gangtise knowledge-base purpose, but the implementation uses access keys, long-term tokens, and a local .authorization file even though the registry credential fields say no credential is required.
ℹ Instruction Scope
The SKILL.md routing language strongly encourages use for reading Gangtise document content, but it remains scoped to the stated knowledge-base search use case and does not show prompt-hijacking behavior.
⚠ Install Mechanism
There is no install spec, but the package includes an update.sh helper that deletes and replaces the installed skill from a remote ZIP without an integrity check or signature.
ℹ Credentials
Network calls to Gangtise APIs and local workspace file writes are expected for this integration, but users should understand that queries, authorization headers, and optional downloads are handled by the scripts.
⚠ Persistence & Privilege
The skill reads a persistent local .authorization file and the updater preserves it across code replacement, giving the skill ongoing access to Gangtise credentials.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install gangtise-knowledgebase
  3. After installation, invoke the skill by name or use /gangtise-knowledgebase
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.4.1
- 更新说明，强调高相关度内容检索，突出语义匹配的专业性和准确性。 - 详细区分了 gangtise-kb 与 gangtise-file 的使用场景与优势。 - 新增文件类型与参数说明，完善会议纪要等分类解释。 - 增加参数表与命令行示例，便于用户快速理解和上手。
Metadata
Slug gangtise-knowledgebase
Version 1.4.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Gangtise知识库?

对 Gangtise 内部知识库（报告、观点、会议纪要等）进行语义检索，返回高相关度的内容片段。当你需要阅读或推理文档的真实内容（如论点、结论、段落等），并且不关心如何定位文件时，务必使用本技能；若你更偏向先按类型/日期/证券等定位筛选再下载核验，请使用 `gangtise-file`。检索结果以语义匹配为核心，... It is an AI Agent Skill for Claude Code / OpenClaw, with 30 downloads so far.

How do I install Gangtise知识库?

Run "/install gangtise-knowledgebase" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Gangtise知识库 free?

Yes, Gangtise知识库 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Gangtise知识库 support?

Gangtise知识库 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Gangtise知识库?

It is built and maintained by GangtiseGts (@gangtisegts); the current version is v1.4.1.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments