← Back to Skills Marketplace
80
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install liqun-context-manager
Description
管理对话上下文和记忆提取。根据关键词高效检索历史记忆。
README (SKILL.md)
Context Manager Skill
功能
- 搜索记忆文件
- 提取相关上下文
- 管理会话状态
使用方法
from context_manager import search_memory, get_recent
# 搜索记忆
results = search_memory('关键词')
# 获取最近记忆
recent = get_recent(5)
Usage Guidance
This skill's code is small and mostly consistent with a memory-search tool, but it also searches a 'skills' directory which is not documented in SKILL.md. That could reveal files belonging to other skills (including configs or secrets) if such files exist. Before installing: (1) ask the author why 'skills' is included and request limiting SEARCH_DIRS to the dedicated memory directory; (2) inspect the contents of any local 'skills' directory you use and remove or relocate sensitive files; (3) consider modifying the code to only return safe metadata (e.g., filenames in a designated memory folder) or to return content snippets rather than full paths; (4) if you run this in an environment with other skill code or secrets, avoid installing until the SEARCH_DIRS behavior is clarified. If you want, I can produce a patched version that only searches the 'memory' directory and/or filters paths by an allowlist.
Capability Analysis
Type: OpenClaw Skill
Name: liqun-context-manager
Version: 1.0.0
The context-manager skill bundle is a straightforward utility for searching and retrieving local markdown and JSON files within designated 'memory' and 'skills' directories. The code in context_manager.py performs basic file I/O and string matching without any network activity, system-level execution, or attempts to access sensitive data outside its defined scope.
Capability Assessment
Purpose & Capability
Name/description (manage conversation context, search memories) align with the provided functions search_memory and get_recent. However, SEARCH_DIRS includes both 'memory' and 'skills' — searching 'skills' is not mentioned in the description and is disproportionate to a memory-only manager.
Instruction Scope
SKILL.md demonstrates use of search_memory and get_recent and makes no mention of scanning a 'skills' directory. The runtime code walks both 'memory' and 'skills' and returns file paths for any .md/.json containing the keyword — this expands scope to potentially read other skill files/configs not related to conversation memory.
Install Mechanism
Instruction-only style with a small included Python module; there is no installer, no downloads, and no packages declared. Low install risk.
Credentials
The skill does not request environment variables, credentials, or config paths. The code does not access environment variables or network resources.
Persistence & Privilege
always is false and no special privileges or persistent system changes are requested. The skill does not modify other skills or system configurations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install liqun-context-manager - After installation, invoke the skill by name or use
/liqun-context-manager - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
liqun-context-manager v1.0.0
- 初始版本发布。
- 支持搜索记忆文件与提取相关对话上下文。
- 提供会话状态管理功能。
- 简单 API：search_memory 和 get_recent 用于检索关键词和获取最近记忆。
Metadata
Frequently Asked Questions
What is LiQun Context Manager?
管理对话上下文和记忆提取。根据关键词高效检索历史记忆。 It is an AI Agent Skill for Claude Code / OpenClaw, with 80 downloads so far.
How do I install LiQun Context Manager?
Run "/install liqun-context-manager" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is LiQun Context Manager free?
Yes, LiQun Context Manager is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does LiQun Context Manager support?
LiQun Context Manager is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created LiQun Context Manager?
It is built and maintained by jaxint (@jaxint); the current version is v1.0.0.
More Skills